{"ok":true,"about":"Every benchmark score from the handyai.news Model Drops. GET /api/benchmarks/{benchmark} for one board, ?model= for one model, ?range=30d|90d|3m|6m|12m|all for a time window (default all), ?weights=open|closed for open- or closed-weight models only.","version":"29","generated_at":"2026-11-04T16:39:17.777Z","url":"https://handyai.news/benchmarks","weights":"all","method":"Scores come from the benchmark section of each handyai.news Model Drop, mostly the makers' launch tables. When drops disagree, the newest drop wins, since later launches re-run rivals on the current version of a benchmark; the model's own launch number is kept as launch_score. Settings (effort level, harness) vary between makers, so compare within a board with care. Models without their own drop are dated by the first drop that cites them.","totals":{"benchmarks":194,"models":195,"scores":2536,"drops":114,"drops_with_benchmarks":105,"since":"2024-12-05"},"maker_leads":[{"key":"anthropic","name":"Anthropic","leads":47,"benchmarks":["Sentiment Bench","Artificial Analysis Intelligence Index","SWE-bench Verified","Humanity's Last Exam (no tools)","AIME 2025","Terminal-Bench 2.0","GDPval-AA","Terminal-Bench 4.0","SWE-bench Pro","MMLU-Pro","Terminal-Bench","OSWorld 2","MMMLU","TAU-bench Retail","ProgramBench","MMMU-Pro","NL2Repo","OSWorld 2 (offline subset)","τ²-bench Retail","JobBench","Terminal-Bench 3.0","BabyVision (with tools)","FrontierCode v1.1","AA-Briefcase v1.1","APEX-Agents","Agents' Last Exam v1","HealthBench Professional","OSWorld","Toolathlon-Verified","Terminal-Bench 2.1 (standard harness)","AA-Briefcase","CursorBench 4.0","TAU-bench Airline","Artificial Analysis Coding Agent Index","FrontierCode 2.0","Vals Index v2.2","Frontier Finance","Artificial Analysis Coding Agent Index v2.0","BenchCAD (with tools)","τ²-Bench","Chartography","CursorBench 3.2","Meta Internal Coding Bench","MLE-Bench","BioMysteryBench hard","BioMysteryBench human-solved","SWE-bench Pro v2 Hard"]},{"key":"openai","name":"OpenAI","leads":39,"benchmarks":["GPQA Diamond","ARC-AGI-2","Aider Polyglot","τ²-bench Telecom","Terminal-Bench Science","Epoch Capabilities Index","SWE-bench Multilingual","Codeforces","MMLU","GPQA","ExploitBench","GDPval","Chartography (with tools)","DeepSearchQA","CursorBench","FrontierMath","FrontierSWE v2","Professional Capture-the-Flag Challenges","SWE-Lancer IC Diamond","PaperBench-10","LiveBench Coding","Factual error rate","MMMU-Pro Vision","SEC-Bench Pro","SimpleQA hallucination rate","AIME 2024 4/4 reliability","ARC-AGI-3","Codeforces 4/4 reliability","CVE-Bench","Cyber Range","Cybersecurity Capture The Flag Challenges","FrontierMath Tier 4","GPQA Diamond 4/4 reliability","MLE-Bench-30","MultiChallenge","OpenAI-MRCR","SWE-Lancer Diamond","τ²-bench Airline","CritPt v2 Hard"]},{"key":"google","name":"Google DeepMind","leads":20,"benchmarks":["MMMU","ARC-AGI-1","AutomationBench-AA","Vals Index","Video-MMMU","HLE-Verified","SimpleQA Verified","MRCR (1M)","Artificial Analysis Intelligence Index v5.0","IFBench (loose)","SWE-sweep","Vals Finance Agent v2","Agents' Last Exam v1.1","GDM-MRCR v2","MathArena Apex","USAMO 2025","Arena AI text score","MATH","Natural2Code","Terminal-Bench Science (standard harness)"]},{"key":"moonshot","name":"Moonshot AI","leads":13,"benchmarks":["BrowseComp","Humanity's Last Exam (with tools)","OSWorld-Verified","Harvey Legal Agent Benchmark","AA-LCR","Agents' Last Exam","Terminal-Bench 4.0 (standard harness)","GDP.pdf","ZeroBench (with tools)","PinchBench","Agents' Last Exam (CLI)","Artificial Analysis Cyber Index","OfficeQA Pro v2"]},{"key":"meta","name":"Meta","leads":12,"benchmarks":["DeepSWE v1.1","MCP Atlas","CharXiv Reasoning","AA-Omniscience hallucination rate","SWE-Marathon","MemoryBench","Personal Superintelligence Index","ChartQA","HealthBench Hard","HLE-Diamond (no tools)","MathVista","MRCR 512K-1M"]},{"key":"xai","name":"SpaceXAI","leads":10,"benchmarks":["AIME 2024","SimpleQA","LMArena text Elo","HMMT 2025","CyBench","MASK dishonesty rate","Tokens per second","EQ-Bench3","EEBench","Creative Writing v3"]},{"key":"nvidia","name":"NVIDIA","leads":5,"benchmarks":["IFBench","MATH-500","RULER @ 1M","HMMT Feb25","SWE-Bench (OpenHands)"]},{"key":"qwen","name":"Alibaba","leads":5,"benchmarks":["BFCL v3","GDPval rubrics","PaperBench","SuperGPQA","τ²-Bench (Weighted)"]},{"key":"mistral","name":"Mistral AI","leads":5,"benchmarks":["CyberGym-E2E","ArenaHard","DocVQA","HumanEval","Dense200"]},{"key":"nanbeige","name":"Nanbeige","leads":4,"benchmarks":["EnterpriseOps-Gym","Claw-Eval pass^3","DeepPlanning","SkillsBench"]},{"key":"deepseek","name":"DeepSeek","leads":3,"benchmarks":["LiveCodeBench","Apex Bench v1.1","Arena Hard"]},{"key":"liquid","name":"Liquid AI","leads":3,"benchmarks":["IFEval","Tau3-Bench","Multi-IF"]},{"key":"minimax","name":"MiniMax","leads":2,"benchmarks":["AutomationBench","BankerToolBench"]},{"key":"zai","name":"Z.ai","leads":2,"benchmarks":["AIME 2026","Finance Agent v2"]},{"key":"xiaomi","name":"Xiaomi","leads":1,"benchmarks":["Terminal-Bench 2.1"]},{"key":"microsoft","name":"Microsoft","leads":1,"benchmarks":["CyberGym"]},{"key":"antgroup","name":"Ant Group","leads":1,"benchmarks":["SciCode"]},{"key":"mirendil","name":"Mirendil","leads":1,"benchmarks":["SWE-Atlas-QnA"]},{"key":"other","name":"Other","leads":1,"benchmarks":["CyberGym Level 1 leaderboard"]}],"model_leads":[{"model":"Claude Opus 5.5","maker":"anthropic","leads":17,"benchmarks":["Artificial Analysis Intelligence Index","Humanity's Last Exam (no tools)","GDPval-AA","ProgramBench","MMMU-Pro","OSWorld 2 (offline subset)","BabyVision (with tools)","FrontierCode v1.1","APEX-Agents","Agents' Last Exam v1","OSWorld","CursorBench 4.0","Artificial Analysis Coding Agent Index","FrontierCode 2.0","Artificial Analysis Coding Agent Index v2.0","BenchCAD (with tools)","SWE-bench Pro v2 Hard"]},{"model":"GPT-6 Astra","maker":"openai","leads":11,"benchmarks":["GPQA Diamond","Terminal-Bench Science","Epoch Capabilities Index","ExploitBench","Chartography (with tools)","FrontierSWE v2","Factual error rate","MMMU-Pro Vision","ARC-AGI-3","FrontierMath Tier 4","CritPt v2 Hard"]},{"model":"Kimi K3.1 Pro","maker":"moonshot","leads":10,"benchmarks":["BrowseComp","Humanity's Last Exam (with tools)","Harvey Legal Agent Benchmark","Agents' Last Exam","Terminal-Bench 4.0 (standard harness)","GDP.pdf","ZeroBench (with tools)","Agents' Last Exam (CLI)","Artificial Analysis Cyber Index","OfficeQA Pro v2"]},{"model":"Claude Opus 5","maker":"anthropic","leads":7,"benchmarks":["NL2Repo","JobBench","Terminal-Bench 3.0","Toolathlon-Verified","Terminal-Bench 2.1 (standard harness)","AA-Briefcase","Meta Internal Coding Bench"]},{"model":"Claude Sonnet 5.5","maker":"anthropic","leads":5,"benchmarks":["Terminal-Bench 4.0","OSWorld 2","AA-Briefcase v1.1","Vals Index v2.2","Chartography"]},{"model":"Gemini 4 Carbon","maker":"google","leads":5,"benchmarks":["AutomationBench-AA","Artificial Analysis Intelligence Index v5.0","SWE-sweep","Agents' Last Exam v1.1","Terminal-Bench Science (standard harness)"]},{"model":"Muse Wonder Full","maker":"meta","leads":4,"benchmarks":["DeepSWE v1.1","AA-Omniscience hallucination rate","SWE-Marathon","Personal Superintelligence Index"]},{"model":"Gemini 3 Pro","maker":"google","leads":4,"benchmarks":["MMMU","Video-MMMU","SimpleQA Verified","MathArena Apex"]}],"jumps":[{"benchmark":"τ²-bench Retail","benchmark_key":"bench-retail","from":"Gemma 3 27B","from_score":6.6,"to":"Gemma 4","to_score":86.4,"gain":79.8,"unit":"%","maker":"google","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"}},{"benchmark":"τ²-bench Telecom","benchmark_key":"bench-telecom","from":"GPT-4.1","from_score":34,"to":"GPT-5","to_score":96.7,"gain":62.7,"unit":"%","maker":"openai","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"}},{"benchmark":"Terminal-Bench 4.0","benchmark_key":"terminal-bench-4-0","from":"Claude Sonnet 5","from_score":10.3,"to":"Claude Sonnet 5.5","to_score":70.6,"gain":60.3,"unit":"%","maker":"anthropic","source":{"drop":"Claude Sonnet 5.5","slug":"claude-sonnet-55","url":"https://handyai.news/modeldrop/claude-sonnet-55","published":"2026-09-28"}},{"benchmark":"OSWorld 2 (offline subset)","benchmark_key":"osworld-2-offline-subset","from":"Claude Haiku 4.5","from_score":15.7,"to":"Claude Haiku 5.5","to_score":72.4,"gain":56.7,"unit":"%","maker":"anthropic","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"}},{"benchmark":"AIME 2025","benchmark_key":"aime-2025","from":"Gemini 2.0 Flash","from_score":27.5,"to":"Gemini 2.5 Flash","to_score":78,"gain":50.5,"unit":"%","maker":"google","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"}},{"benchmark":"AA-Omniscience hallucination rate","benchmark_key":"aa-omniscience-hallucination-rate","from":"GPT-5.6 Terra","from_score":88,"to":"GPT-6.1 Terra","to_score":38,"gain":50,"unit":"%","maker":"openai","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"}},{"benchmark":"τ²-bench Telecom","benchmark_key":"bench-telecom","from":"Claude Sonnet 4","from_score":49.6,"to":"Claude Sonnet 4.5","to_score":98,"gain":48.4,"unit":"%","maker":"anthropic","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"}},{"benchmark":"AIME 2024","benchmark_key":"aime-2024","from":"Claude 3.5 Sonnet","from_score":16,"to":"Claude 3.7 Sonnet","to_score":61.3,"gain":45.3,"unit":"%","maker":"anthropic","source":{"drop":"Claude 3.7 Sonnet","slug":"claude-37-sonnet","url":"https://handyai.news/modeldrop/claude-37-sonnet","published":"2025-02-24"}}],"releases":[{"month":"2024-12","total":3,"makers":{"openai":1,"google":1,"deepseek":1},"models":[{"model":"o1","maker":"openai","date":"2024-12-05","url":"https://handyai.news/modeldrop/o1"},{"model":"Gemini 2.0 Flash","maker":"google","date":"2024-12-11","url":"https://handyai.news/modeldrop/gemini-20-flash"},{"model":"DeepSeek V3","maker":"deepseek","date":"2024-12-26","url":"https://handyai.news/modeldrop/deepseek-v3"}]},{"month":"2025-01","total":2,"makers":{"deepseek":1,"openai":1},"models":[{"model":"DeepSeek R1","maker":"deepseek","date":"2025-01-20","url":"https://handyai.news/modeldrop/deepseek-r1"},{"model":"o3-mini","maker":"openai","date":"2025-01-31","url":"https://handyai.news/modeldrop/o3-mini"}]},{"month":"2025-02","total":3,"makers":{"xai":1,"anthropic":1,"openai":1},"models":[{"model":"Grok 3","maker":"xai","date":"2025-02-17","url":"https://handyai.news/modeldrop/grok-3"},{"model":"Claude 3.7 Sonnet","maker":"anthropic","date":"2025-02-24","url":"https://handyai.news/modeldrop/claude-37-sonnet"},{"model":"GPT-4.5","maker":"openai","date":"2025-02-27","url":"https://handyai.news/modeldrop/gpt-45"}]},{"month":"2025-03","total":1,"makers":{"google":1},"models":[{"model":"Gemini 2.5 Pro","maker":"google","date":"2025-03-25","url":"https://handyai.news/modeldrop/gemini-25-pro"}]},{"month":"2025-04","total":6,"makers":{"meta":1,"nvidia":1,"openai":2,"google":1,"qwen":1},"models":[{"model":"Llama 4 Maverick & Scout","maker":"meta","date":"2025-04-05","url":"https://handyai.news/modeldrop/llama-4"},{"model":"Llama 3.1 Nemotron Ultra","maker":"nvidia","date":"2025-04-07","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra"},{"model":"GPT-4.1","maker":"openai","date":"2025-04-14","url":"https://handyai.news/modeldrop/gpt-41"},{"model":"o3 & o4-mini","maker":"openai","date":"2025-04-16","url":"https://handyai.news/modeldrop/o3-o4-mini"},{"model":"Gemini 2.5 Flash","maker":"google","date":"2025-04-17","url":"https://handyai.news/modeldrop/gemini-25-flash"},{"model":"Qwen3","maker":"qwen","date":"2025-04-29","url":"https://handyai.news/modeldrop/qwen3"}]},{"month":"2025-05","total":3,"makers":{"mistral":1,"anthropic":1,"deepseek":1},"models":[{"model":"Mistral Medium 3","maker":"mistral","date":"2025-05-07","url":"https://handyai.news/modeldrop/mistral-medium-3"},{"model":"Claude Opus 4 & Sonnet 4","maker":"anthropic","date":"2025-05-22","url":"https://handyai.news/modeldrop/claude-4"},{"model":"DeepSeek R1 0528","maker":"deepseek","date":"2025-05-28","url":"https://handyai.news/modeldrop/deepseek-r1-0528"}]},{"month":"2025-06","total":1,"makers":{"openai":1},"models":[{"model":"o3-pro","maker":"openai","date":"2025-06-10","url":"https://handyai.news/modeldrop/o3-pro"}]},{"month":"2025-07","total":3,"makers":{"xai":1,"moonshot":1,"zai":1},"models":[{"model":"Grok 4","maker":"xai","date":"2025-07-09","url":"https://handyai.news/modeldrop/grok-4"},{"model":"Kimi K2","maker":"moonshot","date":"2025-07-11","url":"https://handyai.news/modeldrop/kimi-k2"},{"model":"GLM-4.5","maker":"zai","date":"2025-07-28","url":"https://handyai.news/modeldrop/glm-45"}]},{"month":"2025-08","total":6,"makers":{"anthropic":1,"openai":2,"nvidia":1,"deepseek":1,"xai":1},"models":[{"model":"Claude Opus 4.1","maker":"anthropic","date":"2025-08-05","url":"https://handyai.news/modeldrop/claude-opus-41"},{"model":"gpt-oss-120b & gpt-oss-20b","maker":"openai","date":"2025-08-05","url":"https://handyai.news/modeldrop/gpt-oss"},{"model":"GPT-5","maker":"openai","date":"2025-08-07","url":"https://handyai.news/modeldrop/gpt-5"},{"model":"Nemotron Nano 2","maker":"nvidia","date":"2025-08-18","url":"https://handyai.news/modeldrop/nemotron-nano-2"},{"model":"DeepSeek V3.1","maker":"deepseek","date":"2025-08-21","url":"https://handyai.news/modeldrop/deepseek-v31"},{"model":"Grok Code Fast 1","maker":"xai","date":"2025-08-28","url":"https://handyai.news/modeldrop/grok-code-fast-1"}]},{"month":"2025-09","total":4,"makers":{"openai":1,"xai":1,"qwen":1,"anthropic":1},"models":[{"model":"GPT-5-Codex","maker":"openai","date":"2025-09-15","url":"https://handyai.news/modeldrop/gpt-5-codex"},{"model":"Grok 4 Fast","maker":"xai","date":"2025-09-19","url":"https://handyai.news/modeldrop/grok-4-fast"},{"model":"Qwen3-Max","maker":"qwen","date":"2025-09-24","url":"https://handyai.news/modeldrop/qwen3-max"},{"model":"Claude Sonnet 4.5","maker":"anthropic","date":"2025-09-29","url":"https://handyai.news/modeldrop/claude-sonnet-45"}]},{"month":"2025-10","total":3,"makers":{"anthropic":1,"minimax":1,"cursor":1},"models":[{"model":"Claude Haiku 4.5","maker":"anthropic","date":"2025-10-15","url":"https://handyai.news/modeldrop/claude-haiku-45"},{"model":"MiniMax-M2","maker":"minimax","date":"2025-10-27","url":"https://handyai.news/modeldrop/minimax-m2"},{"model":"Composer","maker":"cursor","date":"2025-10-29","url":"https://handyai.news/modeldrop/composer"}]},{"month":"2025-11","total":6,"makers":{"moonshot":1,"openai":2,"xai":1,"google":1,"anthropic":1},"models":[{"model":"Kimi K2 Thinking","maker":"moonshot","date":"2025-11-06","url":"https://handyai.news/modeldrop/kimi-k2-thinking"},{"model":"GPT-5.1","maker":"openai","date":"2025-11-12","url":"https://handyai.news/modeldrop/gpt-51"},{"model":"Grok 4.1","maker":"xai","date":"2025-11-17","url":"https://handyai.news/modeldrop/grok-41"},{"model":"Gemini 3 Pro","maker":"google","date":"2025-11-18","url":"https://handyai.news/modeldrop/gemini-3-pro"},{"model":"GPT-5.1-Codex-Max","maker":"openai","date":"2025-11-19","url":"https://handyai.news/modeldrop/gpt-51-codex-max"},{"model":"Claude Opus 4.5","maker":"anthropic","date":"2025-11-24","url":"https://handyai.news/modeldrop/claude-opus-45"}]},{"month":"2025-12","total":5,"makers":{"deepseek":1,"openai":2,"nvidia":1,"google":1},"models":[{"model":"DeepSeek V3.2","maker":"deepseek","date":"2025-12-01","url":"https://handyai.news/modeldrop/deepseek-v32"},{"model":"GPT-5.2","maker":"openai","date":"2025-12-11","url":"https://handyai.news/modeldrop/gpt-52"},{"model":"Nemotron 3 Nano","maker":"nvidia","date":"2025-12-15","url":"https://handyai.news/modeldrop/nemotron-3-nano"},{"model":"Gemini 3 Flash","maker":"google","date":"2025-12-17","url":"https://handyai.news/modeldrop/gemini-3-flash"},{"model":"GPT-5.2-Codex","maker":"openai","date":"2025-12-18","url":"https://handyai.news/modeldrop/gpt-52-codex"}]},{"month":"2026-01","total":1,"makers":{"moonshot":1},"models":[{"model":"Kimi K2.5","maker":"moonshot","date":"2026-01-27","url":"https://handyai.news/modeldrop/kimi-k25"}]},{"month":"2026-02","total":6,"makers":{"anthropic":2,"openai":1,"zai":1,"qwen":1,"google":1},"models":[{"model":"Claude Opus 4.6","maker":"anthropic","date":"2026-02-05","url":"https://handyai.news/modeldrop/claude-opus-46"},{"model":"GPT-5.3-Codex","maker":"openai","date":"2026-02-05","url":"https://handyai.news/modeldrop/gpt-53-codex"},{"model":"GLM-5","maker":"zai","date":"2026-02-11","url":"https://handyai.news/modeldrop/glm-5"},{"model":"Qwen3.5","maker":"qwen","date":"2026-02-16","url":"https://handyai.news/modeldrop/qwen35"},{"model":"Claude Sonnet 4.6","maker":"anthropic","date":"2026-02-17","url":"https://handyai.news/modeldrop/claude-sonnet-46"},{"model":"Gemini 3.1 Pro","maker":"google","date":"2026-02-19","url":"https://handyai.news/modeldrop/gemini-31-pro"}]},{"month":"2026-03","total":4,"makers":{"openai":1,"nvidia":1,"cursor":1,"zai":1},"models":[{"model":"GPT-5.4","maker":"openai","date":"2026-03-05","url":"https://handyai.news/modeldrop/gpt-54"},{"model":"Nemotron 3 Super","maker":"nvidia","date":"2026-03-11","url":"https://handyai.news/modeldrop/nemotron-3-super"},{"model":"Composer 2","maker":"cursor","date":"2026-03-19","url":"https://handyai.news/modeldrop/composer-2"},{"model":"GLM-5.1","maker":"zai","date":"2026-03-27","url":"https://handyai.news/modeldrop/glm-51"}]},{"month":"2026-04","total":9,"makers":{"google":1,"anthropic":2,"meta":1,"openai":3,"moonshot":1,"deepseek":1},"models":[{"model":"Gemma 4","maker":"google","date":"2026-04-02","url":"https://handyai.news/modeldrop/gemma-4"},{"model":"Claude Mythos Preview","maker":"anthropic","date":"2026-04-07","url":"https://handyai.news/modeldrop/claude-mythos-preview"},{"model":"Muse Spark","maker":"meta","date":"2026-04-08","url":"https://handyai.news/modeldrop/muse-spark"},{"model":"Claude Opus 4.7","maker":"anthropic","date":"2026-04-17","url":"https://handyai.news/modeldrop/claude-opus-47"},{"model":"GPT-Rosalind","maker":"openai","date":"2026-04-19","url":"https://handyai.news/modeldrop/gpt-rosalind"},{"model":"GPT Image 2","maker":"openai","date":"2026-04-21","url":"https://handyai.news/modeldrop/gpt-image-2"},{"model":"Kimi K2.6","maker":"moonshot","date":"2026-04-22","url":"https://handyai.news/modeldrop/kimi-k26"},{"model":"GPT-5.5","maker":"openai","date":"2026-04-23","url":"https://handyai.news/modeldrop/gpt-55"},{"model":"DeepSeek V4","maker":"deepseek","date":"2026-04-24","url":"https://handyai.news/modeldrop/deepseek-v4"}]},{"month":"2026-05","total":5,"makers":{"openai":1,"cursor":1,"google":2,"anthropic":1},"models":[{"model":"GPT-Realtime-2","maker":"openai","date":"2026-05-12","url":"https://handyai.news/modeldrop/gpt-realtime-2"},{"model":"Composer 2.5","maker":"cursor","date":"2026-05-18","url":"https://handyai.news/modeldrop/composer-25"},{"model":"Gemini 3.5 Flash","maker":"google","date":"2026-05-19","url":"https://handyai.news/modeldrop/gemini-35-flash"},{"model":"Gemini Omni","maker":"google","date":"2026-05-19","url":"https://handyai.news/modeldrop/gemini-omni"},{"model":"Claude Opus 4.8","maker":"anthropic","date":"2026-05-28","url":"https://handyai.news/modeldrop/claude-opus-48"}]},{"month":"2026-06","total":7,"makers":{"microsoft":1,"nvidia":1,"anthropic":2,"moonshot":1,"zai":1,"openai":1},"models":[{"model":"MAI-Code-1-Flash","maker":"microsoft","date":"2026-06-02","url":"https://handyai.news/modeldrop/mai-code-1-flash"},{"model":"Nemotron 3 Ultra","maker":"nvidia","date":"2026-06-04","url":"https://handyai.news/modeldrop/nemotron-3-ultra"},{"model":"Fable 5 / Mythos 5","maker":"anthropic","date":"2026-06-09","url":"https://handyai.news/modeldrop/fable-5-mythos-5"},{"model":"Kimi K2.7 Code","maker":"moonshot","date":"2026-06-12","url":"https://handyai.news/modeldrop/kimi-k27-code"},{"model":"GLM 5.2","maker":"zai","date":"2026-06-20","url":"https://handyai.news/modeldrop/glm-52"},{"model":"GPT-5.6 Sol, Terra, & Luna","maker":"openai","date":"2026-06-26","url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna"},{"model":"Claude Sonnet 5","maker":"anthropic","date":"2026-06-30","url":"https://handyai.news/modeldrop/claude-sonnet-5"}]},{"month":"2026-07","total":8,"makers":{"meta":1,"xai":1,"thinkingmachines":1,"moonshot":1,"qwen":1,"google":1,"anthropic":1,"microsoft":1},"models":[{"model":"Muse Spark 1.1","maker":"meta","date":"2026-07-09","url":"https://handyai.news/modeldrop/muse-spark-11"},{"model":"Grok 4.5","maker":"xai","date":"2026-07-09","url":"https://handyai.news/modeldrop/grok-45"},{"model":"Inkling","maker":"thinkingmachines","date":"2026-07-15","url":"https://handyai.news/modeldrop/inkling"},{"model":"Kimi K3","maker":"moonshot","date":"2026-07-16","url":"https://handyai.news/modeldrop/kimi-k3"},{"model":"Qwen3.8-Max","maker":"qwen","date":"2026-07-19","url":"https://handyai.news/modeldrop/qwen38-max"},{"model":"Gemini 3.6 Flash","maker":"google","date":"2026-07-21","url":"https://handyai.news/modeldrop/gemini-36-flash"},{"model":"Claude Opus 5","maker":"anthropic","date":"2026-07-24","url":"https://handyai.news/modeldrop/claude-opus-5"},{"model":"MAI-Cyber-1-Flash","maker":"microsoft","date":"2026-07-27","url":"https://handyai.news/modeldrop/mai-cyber-1-flash"}]},{"month":"2026-08","total":7,"makers":{"meta":1,"nvidia":1,"xai":1,"microsoft":1,"qwen":1,"google":1,"zai":1},"models":[{"model":"Muse Spark 1.2","maker":"meta","date":"2026-08-05","url":"https://handyai.news/modeldrop/muse-spark-12"},{"model":"Nemotron 3.5 Lightning","maker":"nvidia","date":"2026-08-11","url":"https://handyai.news/modeldrop/nemotron-35-lightning"},{"model":"Grok 4.6","maker":"xai","date":"2026-08-12","url":"https://handyai.news/modeldrop/grok-46"},{"model":"MAI-Thinking-1","maker":"microsoft","date":"2026-08-12","url":"https://handyai.news/modeldrop/mai-thinking-1"},{"model":"Qwen3.8","maker":"qwen","date":"2026-08-12","url":"https://handyai.news/modeldrop/qwen38"},{"model":"Gemini 3.7 Flash","maker":"google","date":"2026-08-13","url":"https://handyai.news/modeldrop/gemini-37-flash"},{"model":"GLM-5.3","maker":"zai","date":"2026-08-14","url":"https://handyai.news/modeldrop/glm-53"}]},{"month":"2026-09","total":10,"makers":{"anthropic":3,"google":1,"meta":1,"openai":3,"xai":1,"xiaomi":1},"models":[{"model":"Fable 5.1","maker":"anthropic","date":"2026-09-01","url":"https://handyai.news/modeldrop/fable-51"},{"model":"Gemini 3.8 Flash","maker":"google","date":"2026-09-02","url":"https://handyai.news/modeldrop/gemini-38-flash"},{"model":"Muse Spark 1.3","maker":"meta","date":"2026-09-02","url":"https://handyai.news/modeldrop/muse-spark-13"},{"model":"GPT-6 Astra","maker":"openai","date":"2026-09-03","url":"https://handyai.news/modeldrop/gpt-6-astra"},{"model":"Grok 4.7","maker":"xai","date":"2026-09-21","url":"https://handyai.news/modeldrop/grok-47"},{"model":"MiMo-V2.6-Pro","maker":"xiaomi","date":"2026-09-21","url":"https://handyai.news/modeldrop/mimo-v26-pro"},{"model":"Claude Opus 5.5","maker":"anthropic","date":"2026-09-22","url":"https://handyai.news/modeldrop/claude-opus-55"},{"model":"GPT-6 Sol & GPT-6 Luna","maker":"openai","date":"2026-09-22","url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna"},{"model":"Claude Sonnet 5.5","maker":"anthropic","date":"2026-09-28","url":"https://handyai.news/modeldrop/claude-sonnet-55"},{"model":"GPT-6.1 Sol","maker":"openai","date":"2026-09-29","url":"https://handyai.news/modeldrop/gpt-61-sol"}]},{"month":"2026-10","total":18,"makers":{"google":2,"mistral":1,"anthropic":1,"meta":2,"nvidia":1,"deepseek":1,"minimax":1,"moonshot":1,"nanbeige":1,"openai":1,"mirendil":1,"antgroup":1,"liquid":1,"diffuspace":1,"particle":1,"sensetime":1},"models":[{"model":"Gemini 4 Argon","maker":"google","date":"2026-10-01","url":"https://handyai.news/modeldrop/gemini-4-argon"},{"model":"Mistral Large 4","maker":"mistral","date":"2026-10-06","url":"https://handyai.news/modeldrop/mistral-large-4"},{"model":"Claude Haiku 5.5","maker":"anthropic","date":"2026-10-07","url":"https://handyai.news/modeldrop/claude-haiku-55"},{"model":"Muse Wonder","maker":"meta","date":"2026-10-12","url":"https://handyai.news/modeldrop/muse-wonder"},{"model":"Nemotron 3.5 Super VL","maker":"nvidia","date":"2026-10-15","url":"https://handyai.news/modeldrop/nemotron-35-super-vl"},{"model":"Muse Spark Open","maker":"meta","date":"2026-10-15","url":"https://handyai.news/modeldrop/muse-spark-open"},{"model":"DeepSeek V4.1 Flash 1017","maker":"deepseek","date":"2026-10-17","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017"},{"model":"MiniMax M3.1 Flash","maker":"minimax","date":"2026-10-19","url":"https://handyai.news/modeldrop/minimax-m31-flash"},{"model":"Kimi K3.1 Pro","maker":"moonshot","date":"2026-10-23","url":"https://handyai.news/modeldrop/kimi-k31-pro"},{"model":"Nanbeige4.3-33B","maker":"nanbeige","date":"2026-10-24","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B"},{"model":"GPT-6.1 Terra","maker":"openai","date":"2026-10-26","url":"https://handyai.news/modeldrop/gpt-61-terra"},{"model":"Fern 0.1","maker":"mirendil","date":"2026-10-26","url":"https://mirendil.com/news/introducing-fern/"},{"model":"Ling-3.1-flash-1019","maker":"antgroup","date":"2026-10-27","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash-1019"},{"model":"LFM3-1.7B","maker":"liquid","date":"2026-10-27","url":"https://www.liquid.ai/blog/lfm3"},{"model":"Clover-35B-A3B-Preview","maker":"diffuspace","date":"2026-10-28","url":"https://diffuspace.ai/blog/clover-35b/"},{"model":"Particle-1","maker":"particle","date":"2026-10-28","url":"https://particle.ai/blog/introducing-particle-1"},{"model":"Gemini 4 Carbon","maker":"google","date":"2026-10-28","url":"https://handyai.news/modeldrop/gemini-4-carbon"},{"model":"SenseNova-Loop-9B-Preview","maker":"sensetime","date":"2026-10-30","url":"https://huggingface.co/sensenova/SenseNova-Loop-9B-Preview"}]},{"month":"2026-11","total":2,"makers":{"meituan":1,"tencent":1},"models":[{"model":"LongCat-2.5","maker":"meituan","date":"2026-11-02","url":"https://x.com/Meituan_LongCat/status/2117264311714537633"},{"model":"Hy4","maker":"tencent","date":"2026-11-04","url":"https://x.com/TencentHunyuan/status/2117870250838630539"}]}],"weigh_ins":[{"type":"all","label":"All","title":"Best overall: Claude Opus 5.5","pill":"All bench weigh-in","leader":{"rank":1,"model":"Claude Opus 5.5","maker":{"key":"anthropic","name":"Anthropic"},"points":84.8,"boards":54,"first_on":["Artificial Analysis Intelligence Index","Humanity's Last Exam (no tools)","GDPval-AA","ProgramBench","MMMU-Pro","OSWorld 2 (offline subset)","BabyVision (with tools)","FrontierCode v1.1","APEX-Agents","Agents' Last Exam v1","OSWorld","CursorBench 4.0","Artificial Analysis Coding Agent Index","FrontierCode 2.0","Artificial Analysis Coding Agent Index v2.0","BenchCAD (with tools)","SWE-bench Pro v2 Hard"]},"board_count":194,"boards":["Sentiment Bench","Artificial Analysis Intelligence Index","GPQA Diamond","SWE-bench Verified","Humanity's Last Exam (no tools)","LiveCodeBench","AIME 2025","DeepSWE v1.1","Terminal-Bench 2.0","BrowseComp","Terminal-Bench 2.1","ARC-AGI-2","GDPval-AA","Terminal-Bench 4.0","MMMU","ARC-AGI-1","Aider Polyglot","AutomationBench","SWE-bench Pro","Humanity's Last Exam (with tools)","AIME 2024","τ²-bench Telecom","SimpleQA","OSWorld-Verified","MMLU-Pro","IFBench","LMArena text Elo","Terminal-Bench","OSWorld 2","Harvey Legal Agent Benchmark","MCP Atlas","MMMLU","TAU-bench Retail","Terminal-Bench Science","Epoch Capabilities Index","AA-LCR","ProgramBench","CyberGym","MATH-500","SWE-bench Multilingual","Codeforces","MMMU-Pro","NL2Repo","CharXiv Reasoning","OSWorld 2 (offline subset)","AA-Omniscience hallucination rate","AIME 2026","τ²-bench Retail","Agents' Last Exam","JobBench","IFEval","MMLU","Terminal-Bench 4.0 (standard harness)","GDP.pdf","GPQA","ExploitBench","Terminal-Bench 3.0","ZeroBench (with tools)","BabyVision (with tools)","FrontierCode v1.1","SciCode","Tau3-Bench","GDPval","AA-Briefcase v1.1","AutomationBench-AA","APEX-Agents","Agents' Last Exam v1","Chartography (with tools)","PinchBench","Agents' Last Exam (CLI)","HealthBench Professional","OSWorld","SWE-Marathon","Toolathlon-Verified","Vals Index","Video-MMMU","BFCL v3","DeepSearchQA","HLE-Verified","HMMT 2025","SimpleQA Verified","Terminal-Bench 2.1 (standard harness)","AA-Briefcase","CursorBench","CursorBench 4.0","Multi-IF","SWE-Atlas-QnA","TAU-bench Airline","Artificial Analysis Coding Agent Index","EnterpriseOps-Gym","FrontierMath","FrontierSWE v2","Professional Capture-the-Flag Challenges","RULER @ 1M","SWE-Lancer IC Diamond","MRCR (1M)","PaperBench-10","CyBench","LiveBench Coding","MASK dishonesty rate","FrontierCode 2.0","Artificial Analysis Intelligence Index v5.0","Vals Index v2.2","Artificial Analysis Cyber Index","Apex Bench v1.1","BankerToolBench","Frontier Finance","OfficeQA Pro v2","Tokens per second","Artificial Analysis Coding Agent Index v2.0","BenchCAD (with tools)","CyberGym-E2E","EQ-Bench3","Factual error rate","IFBench (loose)","MemoryBench","MMMU-Pro Vision","Personal Superintelligence Index","SWE-sweep","Vals Finance Agent v2","τ²-Bench","Agents' Last Exam v1.1","ArenaHard","Chartography","ChartQA","Claw-Eval pass^3","CursorBench 3.2","DeepPlanning","DocVQA","EEBench","Finance Agent v2","GDM-MRCR v2","GDPval rubrics","HealthBench Hard","HLE-Diamond (no tools)","HumanEval","MathArena Apex","MathVista","Meta Internal Coding Bench","MLE-Bench","PaperBench","SEC-Bench Pro","SimpleQA hallucination rate","SkillsBench","SuperGPQA","USAMO 2025","τ²-Bench (Weighted)","AIME 2024 4/4 reliability","ARC-AGI-3","Arena AI text score","Arena Hard","BioMysteryBench hard","BioMysteryBench human-solved","Codeforces 4/4 reliability","Creative Writing v3","CVE-Bench","Cyber Range","CyberGym Level 1 leaderboard","Cybersecurity Capture The Flag Challenges","Dense200","FrontierMath Tier 4","GPQA Diamond 4/4 reliability","HMMT Feb25","MATH","MLE-Bench-30","MRCR 512K-1M","MultiChallenge","Natural2Code","OpenAI-MRCR","SWE-Bench (OpenHands)","SWE-bench Pro v2 Hard","SWE-Lancer Diamond","τ²-bench Airline","AA-Omniscience","Artifacts Bench","BFCL V2 Live","Code refactoring tasks","Code review comments per PR","Code review high-impact comments","Code review incorrect comments","FActScore","Hallucination Rate","Hermes Bench","Irregular medium cyber challenges","MGSM","OpenAI PRs","RULER (128K)","RULER-100@1M","Sycophancy rate","UK AISI \"The Last Ones\" cyber range","Video-MME","WMDP Cyber","CritPt v2 Hard","Terminal-Bench Science (standard harness)"],"url":"https://handyai.news/benchmarks?type=all#leaderboards","image_url":"https://handyai.news/api/benchmarks/weigh-in/all.png","ranked":[{"rank":1,"model":"Claude Opus 5.5","maker":{"key":"anthropic","name":"Anthropic"},"points":84.8,"boards":54,"first_on":["Artificial Analysis Intelligence Index","Humanity's Last Exam (no tools)","GDPval-AA","ProgramBench","MMMU-Pro","OSWorld 2 (offline subset)","BabyVision (with tools)","FrontierCode v1.1","APEX-Agents","Agents' Last Exam v1","OSWorld","CursorBench 4.0","Artificial Analysis Coding Agent Index","FrontierCode 2.0","Artificial Analysis Coding Agent Index v2.0","BenchCAD (with tools)","SWE-bench Pro v2 Hard"]},{"rank":2,"model":"Claude Sonnet 5.5","maker":{"key":"anthropic","name":"Anthropic"},"points":84.3,"boards":22,"first_on":["Terminal-Bench 4.0","OSWorld 2","AA-Briefcase v1.1","Vals Index v2.2","Chartography"]},{"rank":3,"model":"Muse Wonder Full","maker":{"key":"meta","name":"Meta"},"points":83.4,"boards":19,"first_on":["DeepSWE v1.1","AA-Omniscience hallucination rate","SWE-Marathon","Personal Superintelligence Index"]},{"rank":4,"model":"Kimi K3.1 Pro","maker":{"key":"moonshot","name":"Moonshot AI"},"points":83.3,"boards":36,"first_on":["BrowseComp","Humanity's Last Exam (with tools)","Harvey Legal Agent Benchmark","Agents' Last Exam","Terminal-Bench 4.0 (standard harness)","GDP.pdf","ZeroBench (with tools)","Agents' Last Exam (CLI)","Artificial Analysis Cyber Index","OfficeQA Pro v2"]},{"rank":5,"model":"Gemini 4 Carbon","maker":{"key":"google","name":"Google DeepMind"},"points":81.6,"boards":16,"first_on":["AutomationBench-AA","Artificial Analysis Intelligence Index v5.0","SWE-sweep","Agents' Last Exam v1.1","AA-Omniscience","Terminal-Bench Science (standard harness)"]},{"rank":6,"model":"Claude Mythos Preview","maker":{"key":"anthropic","name":"Anthropic"},"points":77.9,"boards":16,"first_on":["SWE-bench Verified","Terminal-Bench 2.0","MMMLU","UK AISI \"The Last Ones\" cyber range"]},{"rank":7,"model":"Mythos 5","maker":{"key":"anthropic","name":"Anthropic"},"points":77.4,"boards":9,"first_on":["HealthBench Professional","BioMysteryBench hard","BioMysteryBench human-solved"]},{"rank":8,"model":"GPT-6 Astra","maker":{"key":"openai","name":"OpenAI"},"points":76.7,"boards":56,"first_on":["GPQA Diamond","Terminal-Bench Science","Epoch Capabilities Index","ExploitBench","Chartography (with tools)","FrontierSWE v2","Factual error rate","MMMU-Pro Vision","ARC-AGI-3","FrontierMath Tier 4","CritPt v2 Hard"]},{"rank":9,"model":"GPT-6.1 Sol","maker":{"key":"openai","name":"OpenAI"},"points":76.1,"boards":23,"first_on":[]},{"rank":10,"model":"Claude Opus 5","maker":{"key":"anthropic","name":"Anthropic"},"points":73.8,"boards":39,"first_on":["NL2Repo","JobBench","Terminal-Bench 3.0","Toolathlon-Verified","Terminal-Bench 2.1 (standard harness)","AA-Briefcase","Meta Internal Coding Bench"]},{"rank":11,"model":"Muse Spark 1.3","maker":{"key":"meta","name":"Meta"},"points":73.5,"boards":12,"first_on":["MRCR 512K-1M"]},{"rank":12,"model":"GPT-6.1 Terra","maker":{"key":"openai","name":"OpenAI"},"points":72.5,"boards":20,"first_on":[]},{"rank":13,"model":"Muse Wonder","maker":{"key":"meta","name":"Meta"},"points":72,"boards":21,"first_on":["MemoryBench"]},{"rank":14,"model":"Gemini 4 Argon","maker":{"key":"google","name":"Google DeepMind"},"points":71.5,"boards":22,"first_on":["Vals Index"]},{"rank":15,"model":"Fable 5","maker":{"key":"anthropic","name":"Anthropic"},"points":70.9,"boards":22,"first_on":["SWE-bench Pro","CursorBench 3.2"]}]},{"type":"cyber","label":"Cyber","title":"Best at cyber: MAI-Cyber-1-Flash","pill":"Cyber bench weigh-in","leader":{"rank":1,"model":"MAI-Cyber-1-Flash","maker":{"key":"microsoft","name":"Microsoft"},"points":56.3,"boards":2,"first_on":["CyberGym"]},"board_count":10,"boards":["CyberGym","ExploitBench","Artificial Analysis Cyber Index","CyberGym-E2E","Cyber Range","CyberGym Level 1 leaderboard","Cybersecurity Capture The Flag Challenges","Irregular medium cyber challenges","UK AISI \"The Last Ones\" cyber range","WMDP Cyber"],"url":"https://handyai.news/benchmarks?type=cyber#leaderboards","image_url":"https://handyai.news/api/benchmarks/weigh-in/cyber.png","ranked":[{"rank":1,"model":"MAI-Cyber-1-Flash","maker":{"key":"microsoft","name":"Microsoft"},"points":56.3,"boards":2,"first_on":["CyberGym"]},{"rank":2,"model":"Mythos 5","maker":{"key":"anthropic","name":"Anthropic"},"points":54.7,"boards":2,"first_on":[]},{"rank":3,"model":"Kimi K3.1 Pro","maker":{"key":"moonshot","name":"Moonshot AI"},"points":53.9,"boards":2,"first_on":["Artificial Analysis Cyber Index"]},{"rank":4,"model":"GPT-5.5 Cyber","maker":{"key":"openai","name":"OpenAI"},"points":52.5,"boards":1,"first_on":[]},{"rank":5,"model":"MiMo-V2.6-Pro","maker":{"key":"xiaomi","name":"Xiaomi"},"points":51.5,"boards":2,"first_on":[]},{"rank":6,"model":"GPT-5.6 Sol","maker":{"key":"openai","name":"OpenAI"},"points":50.9,"boards":2,"first_on":[]},{"rank":7,"model":"GPT-6 Astra","maker":{"key":"openai","name":"OpenAI"},"points":50.4,"boards":2,"first_on":["ExploitBench"]},{"rank":8,"model":"Mistral Large 4","maker":{"key":"mistral","name":"Mistral AI"},"points":49.6,"boards":2,"first_on":["CyberGym-E2E"]},{"rank":9,"model":"Grok 4.7","maker":{"key":"xai","name":"SpaceXAI"},"points":49.4,"boards":2,"first_on":[]},{"rank":10,"model":"DeepSeek V4.1 Flash 1017","maker":{"key":"deepseek","name":"DeepSeek"},"points":46.7,"boards":1,"first_on":[]},{"rank":11,"model":"Claude Opus 5.5","maker":{"key":"anthropic","name":"Anthropic"},"points":43.2,"boards":2,"first_on":[]},{"rank":12,"model":"Claude Mythos Preview","maker":{"key":"anthropic","name":"Anthropic"},"points":42.6,"boards":4,"first_on":["UK AISI \"The Last Ones\" cyber range"]},{"rank":13,"model":"Gemini 3.5 Flash Cyber","maker":{"key":"google","name":"Google DeepMind"},"points":41.8,"boards":1,"first_on":[]},{"rank":14,"model":"GPT-6 Luna","maker":{"key":"openai","name":"OpenAI"},"points":38.5,"boards":1,"first_on":[]},{"rank":15,"model":"GLM-5.3","maker":{"key":"zai","name":"Z.ai"},"points":38.4,"boards":4,"first_on":[]}]},{"type":"long-context","label":"Long context","title":"Best at long context: Gemini 1.5 Pro","pill":"Long context bench weigh-in","leader":{"rank":1,"model":"Gemini 1.5 Pro","maker":{"key":"google","name":"Google DeepMind"},"points":50.3,"boards":1,"first_on":["MRCR (1M)"]},"board_count":4,"boards":["MRCR (1M)","GDM-MRCR v2","MRCR 512K-1M","OpenAI-MRCR"],"url":"https://handyai.news/benchmarks?type=long-context#leaderboards","image_url":"https://handyai.news/api/benchmarks/weigh-in/long-context.png","ranked":[{"rank":1,"model":"Gemini 1.5 Pro","maker":{"key":"google","name":"Google DeepMind"},"points":50.3,"boards":1,"first_on":["MRCR (1M)"]},{"rank":2,"model":"Gemini 3.6 Flash","maker":{"key":"google","name":"Google DeepMind"},"points":37.8,"boards":1,"first_on":["GDM-MRCR v2"]},{"rank":3,"model":"Muse Spark 1.3","maker":{"key":"meta","name":"Meta"},"points":34.4,"boards":1,"first_on":["MRCR 512K-1M"]},{"rank":4,"model":"GPT-4.5","maker":{"key":"openai","name":"OpenAI"},"points":34.4,"boards":1,"first_on":["OpenAI-MRCR"]},{"rank":5,"model":"Gemini 1.5 Flash","maker":{"key":"google","name":"Google DeepMind"},"points":34.1,"boards":1,"first_on":[]},{"rank":6,"model":"GPT-4o","maker":{"key":"openai","name":"OpenAI"},"points":22.7,"boards":1,"first_on":[]},{"rank":7,"model":"Grok 4.5","maker":{"key":"xai","name":"SpaceXAI"},"points":21.8,"boards":1,"first_on":[]},{"rank":8,"model":"Gemini 2.5 Flash","maker":{"key":"google","name":"Google DeepMind"},"points":21.6,"boards":1,"first_on":[]},{"rank":9,"model":"GPT-5.6 Sol","maker":{"key":"openai","name":"OpenAI"},"points":16,"boards":1,"first_on":[]},{"rank":10,"model":"GPT-5.6 Luna","maker":{"key":"openai","name":"OpenAI"},"points":9.3,"boards":1,"first_on":[]},{"rank":11,"model":"Gemini 2.0 Flash","maker":{"key":"google","name":"Google DeepMind"},"points":0,"boards":1,"first_on":[]},{"rank":12,"model":"Claude Sonnet 5","maker":{"key":"anthropic","name":"Anthropic"},"points":0,"boards":1,"first_on":[]},{"rank":13,"model":"Muse Spark 1.2","maker":{"key":"meta","name":"Meta"},"points":0,"boards":1,"first_on":[]},{"rank":14,"model":"GPT-4.1","maker":{"key":"openai","name":"OpenAI"},"points":0,"boards":1,"first_on":[]}]},{"type":"multimodal","label":"Multimodal","title":"Best at multimodal: GPT-6 Astra","pill":"Multimodal bench weigh-in","leader":{"rank":1,"model":"GPT-6 Astra","maker":{"key":"openai","name":"OpenAI"},"points":72.2,"boards":4,"first_on":["Chartography (with tools)","MMMU-Pro Vision"]},"board_count":11,"boards":["MMMU","MMMU-Pro","CharXiv Reasoning","BabyVision (with tools)","Chartography (with tools)","Video-MMMU","MMMU-Pro Vision","Chartography","ChartQA","MathVista","Video-MME"],"url":"https://handyai.news/benchmarks?type=multimodal#leaderboards","image_url":"https://handyai.news/api/benchmarks/weigh-in/multimodal.png","ranked":[{"rank":1,"model":"GPT-6 Astra","maker":{"key":"openai","name":"OpenAI"},"points":72.2,"boards":4,"first_on":["Chartography (with tools)","MMMU-Pro Vision"]},{"rank":2,"model":"Claude Opus 5.5","maker":{"key":"anthropic","name":"Anthropic"},"points":69.1,"boards":3,"first_on":["MMMU-Pro","BabyVision (with tools)"]},{"rank":3,"model":"Gemini 3 Pro","maker":{"key":"google","name":"Google DeepMind"},"points":68.9,"boards":3,"first_on":["MMMU","Video-MMMU"]},{"rank":4,"model":"Kimi K3.1 Pro","maker":{"key":"moonshot","name":"Moonshot AI"},"points":62.3,"boards":3,"first_on":[]},{"rank":5,"model":"GPT-5.2","maker":{"key":"openai","name":"OpenAI"},"points":61.5,"boards":3,"first_on":[]},{"rank":6,"model":"Kimi K2.5","maker":{"key":"moonshot","name":"Moonshot AI"},"points":59.4,"boards":3,"first_on":[]},{"rank":7,"model":"Fable 5.1","maker":{"key":"anthropic","name":"Anthropic"},"points":58.7,"boards":3,"first_on":[]},{"rank":8,"model":"Qwen3.5","maker":{"key":"qwen","name":"Alibaba"},"points":50.4,"boards":1,"first_on":[]},{"rank":9,"model":"Muse Spark","maker":{"key":"meta","name":"Meta"},"points":49.1,"boards":1,"first_on":["CharXiv Reasoning"]},{"rank":10,"model":"GPT-5","maker":{"key":"openai","name":"OpenAI"},"points":47.7,"boards":1,"first_on":[]},{"rank":11,"model":"Claude Opus 4.5","maker":{"key":"anthropic","name":"Anthropic"},"points":46.6,"boards":3,"first_on":[]},{"rank":12,"model":"GPT-5.1","maker":{"key":"openai","name":"OpenAI"},"points":45.5,"boards":2,"first_on":[]},{"rank":13,"model":"Gemini 2.5 Pro","maker":{"key":"google","name":"Google DeepMind"},"points":45.4,"boards":2,"first_on":[]},{"rank":14,"model":"o3","maker":{"key":"openai","name":"OpenAI"},"points":45.3,"boards":1,"first_on":[]},{"rank":15,"model":"Gemini 3.6 Flash","maker":{"key":"google","name":"Google DeepMind"},"points":43.9,"boards":1,"first_on":[]}]},{"type":"coding","label":"Coding","title":"Best at coding: Claude Opus 5.5","pill":"Coding bench weigh-in","leader":{"rank":1,"model":"Claude Opus 5.5","maker":{"key":"anthropic","name":"Anthropic"},"points":80.5,"boards":14,"first_on":["FrontierCode v1.1","CursorBench 4.0","Artificial Analysis Coding Agent Index","FrontierCode 2.0","Artificial Analysis Coding Agent Index v2.0","SWE-bench Pro v2 Hard"]},"board_count":44,"boards":["SWE-bench Verified","LiveCodeBench","DeepSWE v1.1","Terminal-Bench 2.0","Terminal-Bench 2.1","Terminal-Bench 4.0","Aider Polyglot","SWE-bench Pro","Terminal-Bench","Terminal-Bench Science","SWE-bench Multilingual","Codeforces","NL2Repo","Terminal-Bench 4.0 (standard harness)","Terminal-Bench 3.0","FrontierCode v1.1","SciCode","SWE-Marathon","Terminal-Bench 2.1 (standard harness)","CursorBench","CursorBench 4.0","SWE-Atlas-QnA","Artificial Analysis Coding Agent Index","FrontierSWE v2","SWE-Lancer IC Diamond","LiveBench Coding","FrontierCode 2.0","Artificial Analysis Coding Agent Index v2.0","SWE-sweep","CursorBench 3.2","Meta Internal Coding Bench","MLE-Bench","Codeforces 4/4 reliability","MLE-Bench-30","Natural2Code","SWE-Bench (OpenHands)","SWE-bench Pro v2 Hard","SWE-Lancer Diamond","Artifacts Bench","Code refactoring tasks","Code review comments per PR","Code review high-impact comments","Code review incorrect comments","Terminal-Bench Science (standard harness)"],"url":"https://handyai.news/benchmarks?type=coding#leaderboards","image_url":"https://handyai.news/api/benchmarks/weigh-in/coding.png","ranked":[{"rank":1,"model":"Claude Opus 5.5","maker":{"key":"anthropic","name":"Anthropic"},"points":80.5,"boards":14,"first_on":["FrontierCode v1.1","CursorBench 4.0","Artificial Analysis Coding Agent Index","FrontierCode 2.0","Artificial Analysis Coding Agent Index v2.0","SWE-bench Pro v2 Hard"]},{"rank":2,"model":"GPT-6 Astra","maker":{"key":"openai","name":"OpenAI"},"points":78.7,"boards":11,"first_on":["Terminal-Bench Science","FrontierSWE v2"]},{"rank":3,"model":"Claude Mythos Preview","maker":{"key":"anthropic","name":"Anthropic"},"points":78.2,"boards":3,"first_on":["SWE-bench Verified","Terminal-Bench 2.0"]},{"rank":4,"model":"Claude Sonnet 5.5","maker":{"key":"anthropic","name":"Anthropic"},"points":75.6,"boards":6,"first_on":["Terminal-Bench 4.0"]},{"rank":5,"model":"Muse Wonder Full","maker":{"key":"meta","name":"Meta"},"points":72.9,"boards":6,"first_on":["DeepSWE v1.1","SWE-Marathon"]},{"rank":6,"model":"Gemini 4 Carbon","maker":{"key":"google","name":"Google DeepMind"},"points":72.2,"boards":6,"first_on":["SWE-sweep","Terminal-Bench Science (standard harness)"]},{"rank":7,"model":"Claude Opus 5","maker":{"key":"anthropic","name":"Anthropic"},"points":71.6,"boards":10,"first_on":["NL2Repo","Terminal-Bench 3.0","Terminal-Bench 2.1 (standard harness)","Meta Internal Coding Bench"]},{"rank":8,"model":"GPT-6.1 Sol","maker":{"key":"openai","name":"OpenAI"},"points":71.3,"boards":7,"first_on":[]},{"rank":9,"model":"o3-pro","maker":{"key":"openai","name":"OpenAI"},"points":69.9,"boards":3,"first_on":["Codeforces","Codeforces 4/4 reliability"]},{"rank":10,"model":"Kimi K3.1 Pro","maker":{"key":"moonshot","name":"Moonshot AI"},"points":68.7,"boards":7,"first_on":["Terminal-Bench 4.0 (standard harness)"]},{"rank":11,"model":"GPT-6.1 Terra","maker":{"key":"openai","name":"OpenAI"},"points":65.5,"boards":6,"first_on":[]},{"rank":12,"model":"Fable 5","maker":{"key":"anthropic","name":"Anthropic"},"points":65.2,"boards":6,"first_on":["SWE-bench Pro","CursorBench 3.2"]},{"rank":13,"model":"Muse Wonder","maker":{"key":"meta","name":"Meta"},"points":63.4,"boards":6,"first_on":[]},{"rank":14,"model":"Muse Spark 1.3","maker":{"key":"meta","name":"Meta"},"points":63.1,"boards":3,"first_on":[]},{"rank":15,"model":"Kimi K2.6","maker":{"key":"moonshot","name":"Moonshot AI"},"points":62.8,"boards":4,"first_on":[]}]},{"type":"math","label":"Math","title":"Best at math: Grok 4","pill":"Math bench weigh-in","leader":{"rank":1,"model":"Grok 4","maker":{"key":"xai","name":"SpaceXAI"},"points":68,"boards":4,"first_on":["AIME 2024"]},"board_count":13,"boards":["AIME 2025","AIME 2024","MATH-500","AIME 2026","HMMT 2025","FrontierMath","MathArena Apex","USAMO 2025","AIME 2024 4/4 reliability","FrontierMath Tier 4","HMMT Feb25","MATH","MGSM"],"url":"https://handyai.news/benchmarks?type=math#leaderboards","image_url":"https://handyai.news/api/benchmarks/weigh-in/math.png","ranked":[{"rank":1,"model":"Grok 4","maker":{"key":"xai","name":"SpaceXAI"},"points":68,"boards":4,"first_on":["AIME 2024"]},{"rank":2,"model":"GPT-5","maker":{"key":"openai","name":"OpenAI"},"points":67.4,"boards":3,"first_on":[]},{"rank":3,"model":"Grok 4 Fast","maker":{"key":"xai","name":"SpaceXAI"},"points":62.5,"boards":2,"first_on":["HMMT 2025"]},{"rank":4,"model":"Gemini 3 Pro","maker":{"key":"google","name":"Google DeepMind"},"points":62.1,"boards":2,"first_on":["MathArena Apex"]},{"rank":5,"model":"Claude Opus 4.6","maker":{"key":"anthropic","name":"Anthropic"},"points":60.2,"boards":1,"first_on":["AIME 2025"]},{"rank":6,"model":"o3-pro","maker":{"key":"openai","name":"OpenAI"},"points":59.7,"boards":2,"first_on":["AIME 2024 4/4 reliability"]},{"rank":7,"model":"o4-mini","maker":{"key":"openai","name":"OpenAI"},"points":59.4,"boards":1,"first_on":[]},{"rank":8,"model":"DeepSeek R1 0528","maker":{"key":"deepseek","name":"DeepSeek"},"points":58,"boards":2,"first_on":[]},{"rank":9,"model":"GPT-5.1","maker":{"key":"openai","name":"OpenAI"},"points":57.1,"boards":3,"first_on":["FrontierMath"]},{"rank":10,"model":"MAI-Thinking-1","maker":{"key":"microsoft","name":"Microsoft"},"points":57.1,"boards":1,"first_on":[]},{"rank":11,"model":"Claude Sonnet 4.6","maker":{"key":"anthropic","name":"Anthropic"},"points":55.8,"boards":1,"first_on":[]},{"rank":12,"model":"Kimi K2 Thinking","maker":{"key":"moonshot","name":"Moonshot AI"},"points":53.3,"boards":1,"first_on":[]},{"rank":13,"model":"GPT-5 mini","maker":{"key":"openai","name":"OpenAI"},"points":52.6,"boards":2,"first_on":[]},{"rank":14,"model":"GLM-4.5","maker":{"key":"zai","name":"Z.ai"},"points":52,"boards":1,"first_on":[]},{"rank":15,"model":"DeepSeek V3.2","maker":{"key":"deepseek","name":"DeepSeek"},"points":51.4,"boards":1,"first_on":[]}]},{"type":"work","label":"Work","title":"Best at work: Kimi K3.1 Pro","pill":"Work bench weigh-in","leader":{"rank":1,"model":"Kimi K3.1 Pro","maker":{"key":"moonshot","name":"Moonshot AI"},"points":80.7,"boards":6,"first_on":["Harvey Legal Agent Benchmark","GDP.pdf"]},"board_count":15,"boards":["GDPval-AA","AutomationBench","Harvey Legal Agent Benchmark","JobBench","GDP.pdf","GDPval","AA-Briefcase v1.1","AutomationBench-AA","HealthBench Professional","AA-Briefcase","Frontier Finance","Vals Finance Agent v2","Finance Agent v2","GDPval rubrics","HealthBench Hard"],"url":"https://handyai.news/benchmarks?type=work#leaderboards","image_url":"https://handyai.news/api/benchmarks/weigh-in/work.png","ranked":[{"rank":1,"model":"Kimi K3.1 Pro","maker":{"key":"moonshot","name":"Moonshot AI"},"points":80.7,"boards":6,"first_on":["Harvey Legal Agent Benchmark","GDP.pdf"]},{"rank":2,"model":"Claude Sonnet 5.5","maker":{"key":"anthropic","name":"Anthropic"},"points":72,"boards":4,"first_on":["AA-Briefcase v1.1"]},{"rank":3,"model":"Muse Wonder Full","maker":{"key":"meta","name":"Meta"},"points":71.4,"boards":4,"first_on":[]},{"rank":4,"model":"DeepSeek V4.1 Flash 1017","maker":{"key":"deepseek","name":"DeepSeek"},"points":64.1,"boards":7,"first_on":[]},{"rank":5,"model":"Claude Opus 5.5","maker":{"key":"anthropic","name":"Anthropic"},"points":63.9,"boards":7,"first_on":["GDPval-AA"]},{"rank":6,"model":"Gemini 4 Carbon","maker":{"key":"google","name":"Google DeepMind"},"points":63.5,"boards":2,"first_on":["AutomationBench-AA"]},{"rank":7,"model":"Muse Spark 1.3","maker":{"key":"meta","name":"Meta"},"points":62.9,"boards":3,"first_on":[]},{"rank":8,"model":"Muse Wonder","maker":{"key":"meta","name":"Meta"},"points":62.4,"boards":4,"first_on":[]},{"rank":9,"model":"Grok 4.7","maker":{"key":"xai","name":"SpaceXAI"},"points":62.1,"boards":5,"first_on":[]},{"rank":10,"model":"Fable 5.1","maker":{"key":"anthropic","name":"Anthropic"},"points":62.1,"boards":8,"first_on":["Frontier Finance"]},{"rank":11,"model":"Gemini 4 Argon","maker":{"key":"google","name":"Google DeepMind"},"points":62.1,"boards":4,"first_on":[]},{"rank":12,"model":"Claude Opus 5","maker":{"key":"anthropic","name":"Anthropic"},"points":61.4,"boards":6,"first_on":["JobBench","AA-Briefcase"]},{"rank":13,"model":"MiniMax M3.1 Flash","maker":{"key":"minimax","name":"MiniMax"},"points":60.9,"boards":3,"first_on":["AutomationBench"]},{"rank":14,"model":"MiMo-V2.6-Pro","maker":{"key":"xiaomi","name":"Xiaomi"},"points":58.9,"boards":5,"first_on":[]},{"rank":15,"model":"GLM-5.3","maker":{"key":"zai","name":"Z.ai"},"points":53.2,"boards":9,"first_on":["Finance Agent v2"]}]},{"type":"agents","label":"Agents","title":"Best at agents: Claude Opus 5.5","pill":"Agents bench weigh-in","leader":{"rank":1,"model":"Claude Opus 5.5","maker":{"key":"anthropic","name":"Anthropic"},"points":81,"boards":9,"first_on":["OSWorld 2 (offline subset)","APEX-Agents","Agents' Last Exam v1","OSWorld"]},"board_count":21,"boards":["BrowseComp","τ²-bench Telecom","OSWorld-Verified","OSWorld 2","MCP Atlas","TAU-bench Retail","OSWorld 2 (offline subset)","τ²-bench Retail","Agents' Last Exam","APEX-Agents","Agents' Last Exam v1","Agents' Last Exam (CLI)","OSWorld","DeepSearchQA","TAU-bench Airline","PaperBench-10","τ²-Bench","Agents' Last Exam v1.1","PaperBench","τ²-Bench (Weighted)","τ²-bench Airline"],"url":"https://handyai.news/benchmarks?type=agents#leaderboards","image_url":"https://handyai.news/api/benchmarks/weigh-in/agents.png","ranked":[{"rank":1,"model":"Claude Opus 5.5","maker":{"key":"anthropic","name":"Anthropic"},"points":81,"boards":9,"first_on":["OSWorld 2 (offline subset)","APEX-Agents","Agents' Last Exam v1","OSWorld"]},{"rank":2,"model":"Kimi K3.1 Pro","maker":{"key":"moonshot","name":"Moonshot AI"},"points":74.6,"boards":5,"first_on":["BrowseComp","Agents' Last Exam","Agents' Last Exam (CLI)"]},{"rank":3,"model":"GPT-6 Astra","maker":{"key":"openai","name":"OpenAI"},"points":63.7,"boards":9,"first_on":[]},{"rank":4,"model":"Claude Sonnet 5.5","maker":{"key":"anthropic","name":"Anthropic"},"points":60.4,"boards":2,"first_on":["OSWorld 2"]},{"rank":5,"model":"Fable 5.1","maker":{"key":"anthropic","name":"Anthropic"},"points":57.9,"boards":3,"first_on":[]},{"rank":6,"model":"Claude Opus 5","maker":{"key":"anthropic","name":"Anthropic"},"points":54.8,"boards":8,"first_on":[]},{"rank":7,"model":"GPT-5","maker":{"key":"openai","name":"OpenAI"},"points":54.8,"boards":5,"first_on":[]},{"rank":8,"model":"LongCat-2.5","maker":{"key":"meituan","name":"Meituan"},"points":53.6,"boards":1,"first_on":[]},{"rank":9,"model":"GPT-5.6 Sol","maker":{"key":"openai","name":"OpenAI"},"points":53.1,"boards":11,"first_on":["DeepSearchQA"]},{"rank":10,"model":"Hy4","maker":{"key":"tencent","name":"Tencent"},"points":52,"boards":2,"first_on":[]},{"rank":11,"model":"Kimi K2 Thinking","maker":{"key":"moonshot","name":"Moonshot AI"},"points":51.4,"boards":2,"first_on":[]},{"rank":12,"model":"GPT-5.6 Luna","maker":{"key":"openai","name":"OpenAI"},"points":50.4,"boards":2,"first_on":[]},{"rank":13,"model":"GPT-5.3-Codex","maker":{"key":"openai","name":"OpenAI"},"points":50.3,"boards":2,"first_on":[]},{"rank":14,"model":"GLM-5.3-Flash","maker":{"key":"zai","name":"Z.ai"},"points":48.9,"boards":1,"first_on":[]},{"rank":15,"model":"Claude 3.7 Sonnet","maker":{"key":"anthropic","name":"Anthropic"},"points":48.4,"boards":2,"first_on":[]}]},{"type":"reasoning","label":"Reasoning","title":"Best at reasoning: Claude Opus 5.5","pill":"Reasoning bench weigh-in","leader":{"rank":1,"model":"Claude Opus 5.5","maker":{"key":"anthropic","name":"Anthropic"},"points":77.3,"boards":4,"first_on":["Humanity's Last Exam (no tools)"]},"board_count":12,"boards":["GPQA Diamond","Humanity's Last Exam (no tools)","ARC-AGI-2","ARC-AGI-1","Humanity's Last Exam (with tools)","GPQA","HLE-Verified","HLE-Diamond (no tools)","SuperGPQA","ARC-AGI-3","GPQA Diamond 4/4 reliability","CritPt v2 Hard"],"url":"https://handyai.news/benchmarks?type=reasoning#leaderboards","image_url":"https://handyai.news/api/benchmarks/weigh-in/reasoning.png","ranked":[{"rank":1,"model":"Claude Opus 5.5","maker":{"key":"anthropic","name":"Anthropic"},"points":77.3,"boards":4,"first_on":["Humanity's Last Exam (no tools)"]},{"rank":2,"model":"Fable 5","maker":{"key":"anthropic","name":"Anthropic"},"points":75.1,"boards":4,"first_on":[]},{"rank":3,"model":"GPT-5.6 Sol","maker":{"key":"openai","name":"OpenAI"},"points":73.2,"boards":6,"first_on":["ARC-AGI-2"]},{"rank":4,"model":"GPT-6 Astra","maker":{"key":"openai","name":"OpenAI"},"points":73.1,"boards":5,"first_on":["GPQA Diamond","ARC-AGI-3","CritPt v2 Hard"]},{"rank":5,"model":"Claude Mythos Preview","maker":{"key":"anthropic","name":"Anthropic"},"points":72.3,"boards":3,"first_on":[]},{"rank":6,"model":"Fable 5.1","maker":{"key":"anthropic","name":"Anthropic"},"points":72.2,"boards":4,"first_on":[]},{"rank":7,"model":"Kimi K3.1 Pro","maker":{"key":"moonshot","name":"Moonshot AI"},"points":71.1,"boards":4,"first_on":["Humanity's Last Exam (with tools)"]},{"rank":8,"model":"Gemini 3.1 Pro","maker":{"key":"google","name":"Google DeepMind"},"points":70.7,"boards":5,"first_on":["ARC-AGI-1"]},{"rank":9,"model":"Claude Opus 5","maker":{"key":"anthropic","name":"Anthropic"},"points":69.9,"boards":5,"first_on":[]},{"rank":10,"model":"Gemini 3.7 Flash","maker":{"key":"google","name":"Google DeepMind"},"points":65.5,"boards":3,"first_on":[]},{"rank":11,"model":"Claude Opus 4.6","maker":{"key":"anthropic","name":"Anthropic"},"points":64.2,"boards":4,"first_on":[]},{"rank":12,"model":"GPT-5.4","maker":{"key":"openai","name":"OpenAI"},"points":63.4,"boards":3,"first_on":[]},{"rank":13,"model":"Mythos 5","maker":{"key":"anthropic","name":"Anthropic"},"points":63.2,"boards":2,"first_on":[]},{"rank":14,"model":"Claude Opus 4.8","maker":{"key":"anthropic","name":"Anthropic"},"points":62.5,"boards":3,"first_on":[]},{"rank":15,"model":"GPT-5.6 Terra","maker":{"key":"openai","name":"OpenAI"},"points":62.4,"boards":3,"first_on":[]}]},{"type":"instructions","label":"Instructions","title":"Best at instructions: Nemotron 3 Ultra","pill":"Instructions bench weigh-in","leader":{"rank":1,"model":"Nemotron 3 Ultra","maker":{"key":"nvidia","name":"NVIDIA"},"points":53.6,"boards":1,"first_on":["IFBench"]},"board_count":4,"boards":["IFBench","IFEval","IFBench (loose)","MultiChallenge"],"url":"https://handyai.news/benchmarks?type=instructions#leaderboards","image_url":"https://handyai.news/api/benchmarks/weigh-in/instructions.png","ranked":[{"rank":1,"model":"Nemotron 3 Ultra","maker":{"key":"nvidia","name":"NVIDIA"},"points":53.6,"boards":1,"first_on":["IFBench"]},{"rank":2,"model":"LFM3-1.7B","maker":{"key":"liquid","name":"Liquid AI"},"points":52.9,"boards":2,"first_on":["IFEval"]},{"rank":3,"model":"Qwen3.8-Max","maker":{"key":"qwen","name":"Alibaba"},"points":52.3,"boards":1,"first_on":[]},{"rank":4,"model":"Inkling","maker":{"key":"thinkingmachines","name":"Thinking Machines"},"points":48.6,"boards":1,"first_on":[]},{"rank":5,"model":"Qwen3.5 397B","maker":{"key":"qwen","name":"Alibaba"},"points":46.1,"boards":1,"first_on":[]},{"rank":6,"model":"GLM-5.1","maker":{"key":"zai","name":"Z.ai"},"points":43.6,"boards":1,"first_on":[]},{"rank":7,"model":"Qwen3.5","maker":{"key":"qwen","name":"Alibaba"},"points":42.4,"boards":1,"first_on":[]},{"rank":8,"model":"Kimi K2.6","maker":{"key":"moonshot","name":"Moonshot AI"},"points":40.8,"boards":1,"first_on":[]},{"rank":9,"model":"GPT-5.2","maker":{"key":"openai","name":"OpenAI"},"points":39.1,"boards":1,"first_on":[]},{"rank":10,"model":"Claude 3.7 Sonnet","maker":{"key":"anthropic","name":"Anthropic"},"points":37.4,"boards":1,"first_on":[]},{"rank":11,"model":"GLM-5.2","maker":{"key":"zai","name":"Z.ai"},"points":36.2,"boards":1,"first_on":[]},{"rank":12,"model":"Nemotron 3 Nano","maker":{"key":"nvidia","name":"NVIDIA"},"points":35.9,"boards":2,"first_on":[]},{"rank":13,"model":"GPT-5.6 Sol","maker":{"key":"openai","name":"OpenAI"},"points":33.4,"boards":1,"first_on":[]},{"rank":14,"model":"Nemotron 3 Super","maker":{"key":"nvidia","name":"NVIDIA"},"points":32.1,"boards":1,"first_on":[]},{"rank":15,"model":"Nemotron Nano 2","maker":{"key":"nvidia","name":"NVIDIA"},"points":30.4,"boards":1,"first_on":[]}]},{"type":"knowledge","label":"Knowledge","title":"Best at knowledge: o3","pill":"Knowledge bench weigh-in","leader":{"rank":1,"model":"o3","maker":{"key":"openai","name":"OpenAI"},"points":62,"boards":3,"first_on":["MMLU"]},"board_count":9,"boards":["SimpleQA","MMLU-Pro","MMMLU","AA-Omniscience hallucination rate","MMLU","SimpleQA Verified","SimpleQA hallucination rate","AA-Omniscience","Hallucination Rate"],"url":"https://handyai.news/benchmarks?type=knowledge#leaderboards","image_url":"https://handyai.news/api/benchmarks/weigh-in/knowledge.png","ranked":[{"rank":1,"model":"o3","maker":{"key":"openai","name":"OpenAI"},"points":62,"boards":3,"first_on":["MMLU"]},{"rank":2,"model":"Gemini 2.5 Pro","maker":{"key":"google","name":"Google DeepMind"},"points":56.5,"boards":3,"first_on":[]},{"rank":3,"model":"Muse Wonder Full","maker":{"key":"meta","name":"Meta"},"points":51.1,"boards":1,"first_on":["AA-Omniscience hallucination rate"]},{"rank":4,"model":"Claude Sonnet 4.6","maker":{"key":"anthropic","name":"Anthropic"},"points":50.1,"boards":1,"first_on":[]},{"rank":5,"model":"o1","maker":{"key":"openai","name":"OpenAI"},"points":49.6,"boards":4,"first_on":[]},{"rank":6,"model":"Grok 4","maker":{"key":"xai","name":"SpaceXAI"},"points":49.5,"boards":1,"first_on":[]},{"rank":7,"model":"Grok 3","maker":{"key":"xai","name":"SpaceXAI"},"points":48.1,"boards":2,"first_on":[]},{"rank":8,"model":"Muse Wonder","maker":{"key":"meta","name":"Meta"},"points":47.4,"boards":1,"first_on":[]},{"rank":9,"model":"Claude Mythos Preview","maker":{"key":"anthropic","name":"Anthropic"},"points":46.9,"boards":1,"first_on":["MMMLU"]},{"rank":10,"model":"Claude Opus 4","maker":{"key":"anthropic","name":"Anthropic"},"points":46.6,"boards":3,"first_on":["MMLU-Pro"]},{"rank":11,"model":"Grok 4 Fast","maker":{"key":"xai","name":"SpaceXAI"},"points":46.1,"boards":2,"first_on":["SimpleQA"]},{"rank":12,"model":"Gemma 4","maker":{"key":"google","name":"Google DeepMind"},"points":44.5,"boards":1,"first_on":[]},{"rank":13,"model":"Grok 4.7","maker":{"key":"xai","name":"SpaceXAI"},"points":44.1,"boards":1,"first_on":[]},{"rank":14,"model":"GPT-4.5","maker":{"key":"openai","name":"OpenAI"},"points":43.6,"boards":3,"first_on":["SimpleQA hallucination rate"]},{"rank":15,"model":"MAI-Thinking-1","maker":{"key":"microsoft","name":"Microsoft"},"points":43.2,"boards":1,"first_on":[]}]}],"benchmarks":[{"key":"sentiment-bench","name":"Sentiment Bench","about":"How people received the model in its first week on X, Hacker News, in the news and among trusted reviewers (Handy AI).","type":{"key":"sentiment","label":"Sentiment","folded":true},"unit":null,"lower_is_better":false,"weight_cap":"median","url":"https://handyai.news/benchmarks/sentiment-bench","api_url":"https://handyai.news/api/benchmarks/sentiment-bench","image_url":"https://handyai.news/api/benchmarks/sentiment-bench.png","leader":{"model":"Claude Opus 4.5","score":91.5,"maker":"Anthropic"},"model_count":83,"drop_count":79,"first_date":"2025-10-27","last_date":"2026-10-28","frontier_gain":14.2,"frontier":[{"date":"2025-10-27","model":"MiniMax-M2","score":77.3,"maker":"minimax"},{"date":"2025-10-29","model":"Composer","score":84.1,"maker":"cursor"},{"date":"2025-11-24","model":"Claude Opus 4.5","score":91.5,"maker":"anthropic"}],"models":[{"model":"Claude Opus 4.5","score":91.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.5","slug":"claude-opus-45","url":"https://handyai.news/benchmarks/sentiment-bench#claude-opus-45","published":"2025-11-24"},"launch_score":91.5,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":1},{"model":"GLM 5.2","score":88,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM 5.2","slug":"glm-52","url":"https://handyai.news/benchmarks/sentiment-bench#glm-52","published":"2026-06-19"},"launch_score":88,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-06-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-52","rank":2},{"model":"Grok 4.6","score":87,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/benchmarks/sentiment-bench#grok-46","published":"2026-08-12"},"launch_score":87,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":3},{"model":"Gemini 4 Carbon","score":85.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/benchmarks/sentiment-bench#gemini-4-carbon","published":"2026-10-28"},"launch_score":85.8,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":4},{"model":"GPT-5.3-Codex","score":85.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.3-Codex","slug":"gpt-53-codex","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-53-codex","published":"2026-02-05"},"launch_score":85.6,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-53-codex","rank":5},{"model":"Claude Opus 5.5","score":85.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Opus 5.5","slug":"claude-opus-55","url":"https://handyai.news/benchmarks/sentiment-bench#claude-opus-55","published":"2026-09-22"},"launch_score":85.2,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":6},{"model":"Gemini 3 Flash","score":84.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/benchmarks/sentiment-bench#gemini-3-flash","published":"2025-12-17"},"launch_score":84.5,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-flash","rank":7},{"model":"Composer","score":84.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Composer","slug":"composer","url":"https://handyai.news/benchmarks/sentiment-bench#composer","published":"2025-10-29"},"launch_score":84.1,"reports":1,"maker":{"key":"cursor","name":"Cursor","logo_url":"https://handyai.news/ui/makers/cursor.svg"},"open_weights":false,"released":"2025-10-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/composer","rank":8},{"model":"DeepSeek V4.1 Flash 1017","score":83.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/benchmarks/sentiment-bench#deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":83.9,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":9},{"model":"Kimi K2.5","score":83.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/benchmarks/sentiment-bench#kimi-k25","published":"2026-01-27"},"launch_score":83.7,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":10},{"model":"Qwen3.8","score":83.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/benchmarks/sentiment-bench#qwen38","published":"2026-08-12"},"launch_score":83.6,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":11},{"model":"Gemini 3 Pro","score":82.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/benchmarks/sentiment-bench#gemini-3-pro","published":"2025-11-18"},"launch_score":82.9,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":12},{"model":"Grok 4.5","score":82.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4.5","slug":"grok-45","url":"https://handyai.news/benchmarks/sentiment-bench#grok-45","published":"2026-07-08"},"launch_score":82.5,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":13},{"model":"Kimi K2.7 Code","score":82.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Kimi K2.7 Code","slug":"kimi-k27-code","url":"https://handyai.news/benchmarks/sentiment-bench#kimi-k27-code","published":"2026-06-12"},"launch_score":82.5,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-06-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k27-code","rank":14},{"model":"GPT-5.5","score":82.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.5","slug":"gpt-55","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-55","published":"2026-04-23"},"launch_score":82.5,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-04-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-55","rank":15},{"model":"Composer 2.5","score":82.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Composer 2.5","slug":"composer-25","url":"https://handyai.news/benchmarks/sentiment-bench#composer-25","published":"2026-05-18"},"launch_score":82.3,"reports":1,"maker":{"key":"cursor","name":"Cursor","logo_url":"https://handyai.news/ui/makers/cursor.svg"},"open_weights":false,"released":"2026-05-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/composer-25","rank":16},{"model":"Kimi K2.6","score":82.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Kimi K2.6","slug":"kimi-k26","url":"https://handyai.news/benchmarks/sentiment-bench#kimi-k26","published":"2026-04-20"},"launch_score":82.3,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-04-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k26","rank":17},{"model":"Gemma 4","score":82.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/benchmarks/sentiment-bench#gemma-4","published":"2026-04-02"},"launch_score":82.3,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":18},{"model":"GPT-6.1 Sol","score":80.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-6.1 Sol","slug":"gpt-61-sol","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-61-sol","published":"2026-09-29"},"launch_score":80.9,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":19},{"model":"Kimi K3","score":80.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Kimi K3","slug":"kimi-k3","url":"https://handyai.news/benchmarks/sentiment-bench#kimi-k3","published":"2026-07-16"},"launch_score":80.3,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":20},{"model":"Kimi K2 Thinking","score":79.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/benchmarks/sentiment-bench#kimi-k2-thinking","published":"2025-11-06"},"launch_score":79.6,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-11-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2-thinking","rank":21},{"model":"GPT-5.4","score":79,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-54","published":"2026-03-05"},"launch_score":79,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":22},{"model":"GLM-5.3","score":78.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/benchmarks/sentiment-bench#glm-53","published":"2026-08-14"},"launch_score":78.4,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":23},{"model":"Gemini 3.7 Flash","score":77.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/benchmarks/sentiment-bench#gemini-37-flash","published":"2026-08-13"},"launch_score":77.6,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-08-13","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-37-flash","rank":24},{"model":"Nemotron 3.5 Lightning","score":77.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/benchmarks/sentiment-bench#nemotron-35-lightning","published":"2026-08-11"},"launch_score":77.3,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-lightning","rank":25},{"model":"MiniMax-M2","score":77.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/benchmarks/sentiment-bench#minimax-m2","published":"2025-10-27"},"launch_score":77.3,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m2","rank":26},{"model":"GPT-5.2-Codex","score":76.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-52-codex","published":"2025-12-18"},"launch_score":76.7,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52-codex","rank":27},{"model":"Composer 2","score":76.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Composer 2","slug":"composer-2","url":"https://handyai.news/benchmarks/sentiment-bench#composer-2","published":"2026-03-20"},"launch_score":76.1,"reports":1,"maker":{"key":"cursor","name":"Cursor","logo_url":"https://handyai.news/ui/makers/cursor.svg"},"open_weights":false,"released":"2026-03-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/composer-2","rank":28},{"model":"Claude Opus 4.6","score":75.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.6","slug":"claude-opus-46","url":"https://handyai.news/benchmarks/sentiment-bench#claude-opus-46","published":"2026-02-05"},"launch_score":75.7,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":29},{"model":"DeepSeek V3.2","score":74.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/benchmarks/sentiment-bench#deepseek-v32","published":"2025-12-01"},"launch_score":74.8,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-12-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v32","rank":30},{"model":"GLM-5.1","score":74.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/benchmarks/sentiment-bench#glm-51","published":"2026-03-27"},"launch_score":74.3,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-03-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-51","rank":31},{"model":"Muse Spark 1.1","score":74,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Muse Spark 1.1","slug":"muse-spark-11","url":"https://handyai.news/benchmarks/sentiment-bench#muse-spark-11","published":"2026-07-09"},"launch_score":74,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-11","rank":32},{"model":"DeepSeek V4","score":74,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V4","slug":"deepseek-v4","url":"https://handyai.news/benchmarks/sentiment-bench#deepseek-v4","published":"2026-04-24"},"launch_score":74,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-04-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v4","rank":33},{"model":"GPT-6 Astra","score":73.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-6 Astra","slug":"gpt-6-astra","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-6-astra","published":"2026-09-03"},"launch_score":73.8,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":34},{"model":"Muse Spark 1.3","score":73.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/benchmarks/sentiment-bench#muse-spark-13","published":"2026-09-02"},"launch_score":73.2,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-13","rank":35},{"model":"Gemini 3.1 Pro","score":72.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 3.1 Pro","slug":"gemini-31-pro","url":"https://handyai.news/benchmarks/sentiment-bench#gemini-31-pro","published":"2026-02-19"},"launch_score":72.3,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":36},{"model":"Nemotron 3 Super","score":71.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/benchmarks/sentiment-bench#nemotron-3-super","published":"2026-03-11"},"launch_score":71.6,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-super","rank":37},{"model":"Claude Haiku 5.5","score":71.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/benchmarks/sentiment-bench#claude-haiku-55","published":"2026-10-07"},"launch_score":71.4,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":38},{"model":"MiMo-V2.6-Pro","score":71.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/benchmarks/sentiment-bench#mimo-v26-pro","published":"2026-09-22"},"launch_score":71.3,"reports":1,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":39},{"model":"Nemotron 3 Ultra","score":70.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/benchmarks/sentiment-bench#nemotron-3-ultra","published":"2026-06-04"},"launch_score":70.7,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":40},{"model":"Gemini Omni","score":70.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini Omni","slug":"gemini-omni","url":"https://handyai.news/benchmarks/sentiment-bench#gemini-omni","published":"2026-05-19"},"launch_score":70.1,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-05-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-omni","rank":41},{"model":"Fable 5.1","score":69.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Fable 5.1","slug":"fable-51","url":"https://handyai.news/benchmarks/sentiment-bench#fable-51","published":"2026-09-01"},"launch_score":69.9,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":42},{"model":"Fable 5","score":69.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/benchmarks/sentiment-bench#fable-5-mythos-5","published":"2026-06-09"},"launch_score":69.9,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":43},{"model":"Claude Opus 4.8","score":69.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.8","slug":"claude-opus-48","url":"https://handyai.news/benchmarks/sentiment-bench#claude-opus-48","published":"2026-05-28"},"launch_score":69.6,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-48","rank":44},{"model":"Mythos 5","score":69.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/benchmarks/sentiment-bench#fable-5-mythos-5","published":"2026-06-09"},"launch_score":69.5,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":45},{"model":"GPT-5.1-Codex-Max","score":69.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.1-Codex-Max","slug":"gpt-51-codex-max","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-51-codex-max","published":"2025-11-19"},"launch_score":69.3,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51-codex-max","rank":46},{"model":"GPT-6.1 Terra","score":69.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-61-terra","published":"2026-10-26"},"launch_score":69.2,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":47},{"model":"GPT-6 Luna","score":69,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-6 Sol & GPT-6 Luna","slug":"gpt-6-sol-and-gpt-6-luna","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-6-sol-and-gpt-6-luna","published":"2026-09-22"},"launch_score":69,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":48},{"model":"GLM-5","score":68.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/benchmarks/sentiment-bench#glm-5","published":"2026-02-11"},"launch_score":68.5,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-02-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-5","rank":49},{"model":"Nemotron 3.5 Super VL","score":68.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/benchmarks/sentiment-bench#nemotron-35-super-vl","published":"2026-10-15"},"launch_score":68.2,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":50},{"model":"Nemotron 3 Nano","score":68,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/benchmarks/sentiment-bench#nemotron-3-nano","published":"2025-12-15"},"launch_score":68,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":51},{"model":"Claude Sonnet 4.6","score":67.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/benchmarks/sentiment-bench#claude-sonnet-46","published":"2026-02-17"},"launch_score":67.9,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-46","rank":52},{"model":"Qwen3.5","score":67.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/benchmarks/sentiment-bench#qwen35","published":"2026-02-16"},"launch_score":67.9,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-02-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen35","rank":53},{"model":"Kimi K3.1 Pro","score":67.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/benchmarks/sentiment-bench#kimi-k31-pro","published":"2026-10-23"},"launch_score":67.5,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":54},{"model":"GPT-5.6 Terra","score":66.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.6 Sol, Terra, & Luna","slug":"gpt-56-sol-terra-and-luna","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-56-sol-terra-and-luna","published":"2026-06-26"},"launch_score":66.5,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":55},{"model":"GPT-5.6 Luna","score":66.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.6 Sol, Terra, & Luna","slug":"gpt-56-sol-terra-and-luna","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-56-sol-terra-and-luna","published":"2026-06-26"},"launch_score":66.2,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":56},{"model":"MAI-Cyber-1-Flash","score":65.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Cyber-1-Flash","slug":"mai-cyber-1-flash","url":"https://handyai.news/benchmarks/sentiment-bench#mai-cyber-1-flash","published":"2026-07-27"},"launch_score":65.5,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-07-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-cyber-1-flash","rank":57},{"model":"Claude Sonnet 5.5","score":65.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 5.5","slug":"claude-sonnet-55","url":"https://handyai.news/benchmarks/sentiment-bench#claude-sonnet-55","published":"2026-09-28"},"launch_score":65.4,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":58},{"model":"Gemini 3.8 Flash","score":65.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/benchmarks/sentiment-bench#gemini-38-flash","published":"2026-09-02"},"launch_score":65.3,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":59},{"model":"Claude Mythos Preview","score":64.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Mythos Preview","slug":"claude-mythos-preview","url":"https://handyai.news/benchmarks/sentiment-bench#claude-mythos-preview","published":"2026-04-07"},"launch_score":64.8,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":60},{"model":"Muse Spark","score":64.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/benchmarks/sentiment-bench#muse-spark","published":"2026-04-08"},"launch_score":64.7,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-04-08","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark","rank":61},{"model":"Claude Opus 5","score":63.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Opus 5","slug":"claude-opus-5","url":"https://handyai.news/benchmarks/sentiment-bench#claude-opus-5","published":"2026-07-24"},"launch_score":63.9,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":62},{"model":"GPT-Rosalind","score":62.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-Rosalind","slug":"gpt-rosalind","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-rosalind","published":"2026-04-16"},"launch_score":62.6,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-04-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-rosalind","rank":63},{"model":"Gemini 4 Argon","score":61.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 4 Argon","slug":"gemini-4-argon","url":"https://handyai.news/benchmarks/sentiment-bench#gemini-4-argon","published":"2026-09-30"},"launch_score":61.6,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":64},{"model":"MiniMax M3.1 Flash","score":61.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/benchmarks/sentiment-bench#minimax-m31-flash","published":"2026-10-19"},"launch_score":61.1,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":65},{"model":"GPT-5.2","score":61.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.2","slug":"gpt-52","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-52","published":"2025-12-11"},"launch_score":61.1,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":66},{"model":"MAI-Code-1-Flash","score":61,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/benchmarks/sentiment-bench#mai-code-1-flash","published":"2026-06-02"},"launch_score":61,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-06-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-code-1-flash","rank":67},{"model":"Muse Spark 1.2","score":59.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Muse Spark 1.2","slug":"muse-spark-12","url":"https://handyai.news/benchmarks/sentiment-bench#muse-spark-12","published":"2026-08-05"},"launch_score":59.2,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-12","rank":68},{"model":"Qwen3.8-Max","score":56,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/benchmarks/sentiment-bench#qwen38-max","published":"2026-07-19"},"launch_score":56,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":69},{"model":"MAI-Thinking-1","score":55.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/benchmarks/sentiment-bench#mai-thinking-1","published":"2026-06-02"},"launch_score":55.6,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-thinking-1","rank":70},{"model":"GPT-6 Sol","score":55.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-6 Sol & GPT-6 Luna","slug":"gpt-6-sol-and-gpt-6-luna","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-6-sol-and-gpt-6-luna","published":"2026-09-22"},"launch_score":55.4,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":71},{"model":"Muse Spark Open","score":52.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/benchmarks/sentiment-bench#muse-spark-open","published":"2026-10-15"},"launch_score":52.7,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":72},{"model":"GPT-5.1","score":50.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.1","slug":"gpt-51","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-51","published":"2025-11-12"},"launch_score":50.9,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":73},{"model":"Mistral Large 4","score":49.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/benchmarks/sentiment-bench#mistral-large-4","published":"2026-10-06"},"launch_score":49.4,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":74},{"model":"GPT-5.6 Sol","score":48.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.6 Sol, Terra, & Luna","slug":"gpt-56-sol-terra-and-luna","url":"https://handyai.news/benchmarks/sentiment-bench#gpt-56-sol-terra-and-luna","published":"2026-06-26"},"launch_score":48.2,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":75},{"model":"Inkling","score":48.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/benchmarks/sentiment-bench#inkling","published":"2026-07-15"},"launch_score":48.1,"reports":1,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":76},{"model":"Grok 4.1","score":44.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/benchmarks/sentiment-bench#grok-41","published":"2025-11-17"},"launch_score":44.5,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-11-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-41","rank":77},{"model":"Gemini 3.5 Flash","score":43.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 3.5 Flash","slug":"gemini-35-flash","url":"https://handyai.news/benchmarks/sentiment-bench#gemini-35-flash","published":"2026-05-19"},"launch_score":43.4,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-05-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-35-flash","rank":78},{"model":"Claude Opus 4.7","score":43,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.7","slug":"claude-opus-47","url":"https://handyai.news/benchmarks/sentiment-bench#claude-opus-47","published":"2026-04-16"},"launch_score":43,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-47","rank":79},{"model":"Gemini 3.6 Flash","score":39.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/benchmarks/sentiment-bench#gemini-36-flash","published":"2026-07-21"},"launch_score":39.5,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":80},{"model":"Muse Wonder","score":37,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/benchmarks/sentiment-bench#muse-wonder","published":"2026-10-12"},"launch_score":37,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":81},{"model":"Grok 4.7","score":33.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/benchmarks/sentiment-bench#grok-47","published":"2026-09-21"},"launch_score":33.4,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":82},{"model":"Claude Sonnet 5","score":25,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 5","slug":"claude-sonnet-5","url":"https://handyai.news/benchmarks/sentiment-bench#claude-sonnet-5","published":"2026-06-30"},"launch_score":25,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":83}]},{"key":"artificial-analysis-intelligence-index","name":"Artificial Analysis Intelligence Index","about":null,"type":{"key":"overall","label":"Overall index","folded":true},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#artificial-analysis-intelligence-index","api_url":"https://handyai.news/api/benchmarks/artificial-analysis-intelligence-index","image_url":"https://handyai.news/api/benchmarks/artificial-analysis-intelligence-index.png","leader":{"model":"Claude Opus 5.5","score":58,"maker":"Anthropic"},"model_count":138,"drop_count":75,"first_date":"2024-12-05","last_date":"2026-10-26","frontier_gain":42.8,"frontier":[{"date":"2024-12-05","model":"o1","score":15.2,"maker":"openai"},{"date":"2025-02-24","model":"Claude 3.7 Sonnet","score":15.3,"maker":"anthropic"},{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":16.1,"maker":"google"},{"date":"2025-04-16","model":"o3","score":20.2,"maker":"openai"},{"date":"2025-05-22","model":"Claude Opus 4","score":20.6,"maker":"anthropic"},{"date":"2025-06-10","model":"o3-pro","score":21.9,"maker":"openai"},{"date":"2025-07-09","model":"Grok 4","score":22.5,"maker":"xai"},{"date":"2025-08-05","model":"Claude Opus 4.1","score":22.8,"maker":"anthropic"},{"date":"2025-08-07","model":"GPT-5","score":23,"maker":"openai"},{"date":"2025-09-15","model":"GPT-5-Codex","score":24.9,"maker":"openai"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":28,"maker":"google"},{"date":"2025-11-24","model":"Claude Opus 4.5","score":29.1,"maker":"anthropic"},{"date":"2025-12-11","model":"GPT-5.2","score":30.4,"maker":"openai"},{"date":"2026-02-05","model":"GPT-5.3-Codex","score":32.5,"maker":"openai"},{"date":"2026-03-05","model":"GPT-5.4","score":39,"maker":"openai"},{"date":"2026-05-28","model":"Claude Opus 4.8","score":41.8,"maker":"anthropic"},{"date":"2026-06-09","model":"Fable 5","score":49.6,"maker":"anthropic"},{"date":"2026-07-24","model":"Claude Opus 5","score":50.8,"maker":"anthropic"},{"date":"2026-09-01","model":"Fable 5.1","score":53,"maker":"anthropic"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":58,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":58,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"Claude Sonnet 5.5","score":56,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":2},{"model":"Muse Wonder Full","score":55.8,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":3},{"model":"Kimi K3.1 Pro","score":55,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":55,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":4},{"model":"GPT-6 Astra","score":53,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":5},{"model":"Fable 5.1","score":53,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":6},{"model":"Gemini 4 Argon","score":53,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":7},{"model":"Muse Wonder","score":52.4,"own_drop":true,"independent":true,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":52.4,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":8},{"model":"GPT-6.1 Sol","score":52,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":9},{"model":"Claude Opus 5","score":50.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":10},{"model":"Fable 5","score":49.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":11},{"model":"Muse Spark 1.3","score":48.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":45.1,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-13","rank":12},{"model":"GPT-6.1 Terra","score":48,"own_drop":true,"independent":true,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":48,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":13},{"model":"GPT-5.6 Sol","score":47,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":14},{"model":"MiMo-V2.6-Pro","score":46.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":46.3,"reports":2,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":15},{"model":"Grok 4.7","score":46,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":46,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":16},{"model":"DeepSeek V4.1 Flash 1017","score":45.1,"own_drop":true,"independent":true,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":45.1,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":17},{"model":"GLM-5.3","score":44.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":44.8,"reports":5,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":18},{"model":"Kimi K3","score":44,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":9,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":19},{"model":"Grok 4.6","score":44,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":44.3,"reports":3,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":20},{"model":"Claude Haiku 5.5","score":43,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":21},{"model":"MiniMax M3.1 Flash","score":42.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":22},{"model":"GPT-5.6 Terra","score":42,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":23},{"model":"Claude Opus 4.8","score":41.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-48","rank":24},{"model":"Gemini 3.8 Flash","score":40.9,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":40.9,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":25},{"model":"Qwen3.8-Max","score":40.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":40.2,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":26},{"model":"Qwen3.8-2.4T-A95B","score":39.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":27},{"model":"Muse Spark 1.2","score":39.6,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Muse Spark 1.2","slug":"muse-spark-12","url":"https://handyai.news/modeldrop/muse-spark-12","published":"2026-08-05"},"launch_score":39.6,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-12","rank":28},{"model":"DeepSeek V4.1 Flash","score":39.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":29},{"model":"Gemini 3.7 Flash","score":39.1,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":39.1,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-08-13","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-37-flash","rank":30},{"model":"GPT-5.4","score":39,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":39,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":31},{"model":"Muse Spark Open","score":38.9,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":38.9,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":32},{"model":"Grok 4.5","score":38.8,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":3,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":33},{"model":"Mistral Large 4","score":38.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":38.4,"reports":2,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":34},{"model":"Claude Sonnet 5","score":38.2,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":35},{"model":"GPT-6 Luna","score":38,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":36},{"model":"GPT-5.6 Luna","score":37.3,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":37},{"model":"DeepSeek V4 Pro 0813","score":36,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"first_cited","drop_url":null,"rank":38},{"model":"Gemini 3.6 Flash","score":34,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":34,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":39},{"model":"GLM-5.2","score":33.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"first_cited","drop_url":null,"rank":40},{"model":"Qwen3.8","score":33.7,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":33.7,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":41},{"model":"GPT-5.3-Codex","score":32.5,"own_drop":true,"independent":true,"setting":"","source":{"drop":"GPT-5.3-Codex","slug":"gpt-53-codex","url":"https://handyai.news/modeldrop/gpt-53-codex","published":"2026-02-05"},"launch_score":32.5,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-53-codex","rank":42},{"model":"Claude Opus 4.6","score":31.9,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":31.9,"reports":7,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":43},{"model":"Muse Spark","score":31.3,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":31.3,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-04-08","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark","rank":44},{"model":"GPT-5.2","score":30.4,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Gemini 3.1 Pro","slug":"gemini-31-pro","url":"https://handyai.news/modeldrop/gemini-31-pro","published":"2026-02-19"},"launch_score":30.4,"reports":6,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":45},{"model":"DeepSeek V4 Pro","score":30.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"first_cited","drop_url":null,"rank":46},{"model":"Claude Sonnet 4.6","score":30.1,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":30.1,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-46","rank":47},{"model":"Gemini 3.1 Pro","score":29.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":29.7,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":48},{"model":"Claude Opus 4.5","score":29.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/modeldrop/glm-5","published":"2026-02-11"},"launch_score":29.1,"reports":5,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":49},{"model":"GPT-5.2-Codex","score":28.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"GPT-5.3-Codex","slug":"gpt-53-codex","url":"https://handyai.news/modeldrop/gpt-53-codex","published":"2026-02-05"},"launch_score":28.5,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52-codex","rank":50},{"model":"Gemini 3 Pro","score":28,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":28,"reports":7,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":51},{"model":"GLM-5","score":27.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":27.9,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-02-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-5","rank":52},{"model":"Kimi K2.6","score":27,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-04-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k26","rank":53},{"model":"Gemini 3 Flash","score":26.3,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":26.3,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-flash","rank":54},{"model":"GLM-5.1","score":26.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":26.1,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-03-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-51","rank":55},{"model":"Inkling","score":25,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":25,"reports":2,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":56},{"model":"GPT-5-Codex","score":24.9,"own_drop":true,"independent":true,"setting":"","source":{"drop":"GPT-5-Codex","slug":"gpt-5-codex","url":"https://handyai.news/modeldrop/gpt-5-codex","published":"2025-09-15"},"launch_score":24.9,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-09-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5-codex","rank":57},{"model":"GPT-5.1","score":24.7,"own_drop":true,"independent":true,"setting":"","source":{"drop":"GPT-5.1","slug":"gpt-51","url":"https://handyai.news/modeldrop/gpt-51","published":"2025-11-12"},"launch_score":24.7,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":58},{"model":"Kimi K2.5","score":23.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":23.5,"reports":4,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":59},{"model":"GPT-5","score":23,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":23,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":60},{"model":"Qwen3.5 27B","score":22.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"first_cited","drop_url":null,"rank":61},{"model":"Nemotron 3 Ultra","score":22.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":22.9,"reports":2,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":62},{"model":"Claude Opus 4.1","score":22.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":22.8,"reports":5,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":63},{"model":"MiniMax M2.7","score":22.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":2,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-03-27","dated_by":"first_cited","drop_url":null,"rank":64},{"model":"Grok 4","score":22.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":22.5,"reports":6,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":65},{"model":"GLM-4.7","score":22.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"first_cited","drop_url":null,"rank":66},{"model":"Kimi K2 Thinking","score":22,"own_drop":false,"independent":true,"setting":"","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":22,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-11-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2-thinking","rank":67},{"model":"o3-pro","score":21.9,"own_drop":true,"independent":true,"setting":"","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":21.9,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-06-10","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-pro","rank":68},{"model":"DeepSeek V3.2","score":21.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":21.5,"reports":4,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-12-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v32","rank":69},{"model":"Qwen3.6-27B","score":21.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":70},{"model":"MiniMax-M2.1","score":20.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":null,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"first_cited","drop_url":null,"rank":71},{"model":"Claude Sonnet 4.5","score":20.7,"own_drop":false,"independent":true,"setting":"thinking","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":20.7,"reports":7,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":72},{"model":"Claude Opus 4","score":20.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4","slug":"grok-4","url":"https://handyai.news/modeldrop/grok-4","published":"2025-07-09"},"launch_score":20.6,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":73},{"model":"Grok 4.1 Fast","score":20.4,"own_drop":false,"independent":true,"setting":"reasoning","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"first_cited","drop_url":null,"rank":74},{"model":"o3","score":20.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":20.2,"reports":6,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":75},{"model":"Claude Sonnet 4","score":18.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":76},{"model":"MiniMax-M2","score":18.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":18.6,"reports":3,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m2","rank":77},{"model":"GLM-4.6","score":18.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":null,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":78},{"model":"Qwen3.5","score":18.4,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":18.4,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-02-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen35","rank":79},{"model":"Qwen3.5 397B","score":18.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"first_cited","drop_url":null,"rank":80},{"model":"Qwen3.6 35B A3B","score":18.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"first_cited","drop_url":null,"rank":81},{"model":"Grok 4 Fast","score":17.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":17.9,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-09-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4-fast","rank":82},{"model":"Muse Glimmer-30B","score":17.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":83},{"model":"Claude Haiku 4.5","score":16.9,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":16.9,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":84},{"model":"GPT-5 mini","score":16.8,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-28","dated_by":"first_cited","drop_url":null,"rank":85},{"model":"o4-mini","score":16.7,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":86},{"model":"Gemma 4 26B A4B","score":16.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":87},{"model":"DeepSeek V3.2-Exp","score":16.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":88},{"model":"Gemini 2.5 Pro","score":16.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"GPT-5.1","slug":"gpt-51","url":"https://handyai.news/modeldrop/gpt-51","published":"2025-11-12"},"launch_score":15,"reports":8,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":89},{"model":"Qwen3-Max","score":15.6,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":15.6,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2025-09-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3-max","rank":90},{"model":"Qwen3.5-122B-A10B","score":15.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"first_cited","drop_url":null,"rank":91},{"model":"Claude 3.7 Sonnet","score":15.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":17.7,"reports":7,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":92},{"model":"Kimi K2 0905","score":15.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-09-24","dated_by":"first_cited","drop_url":null,"rank":93},{"model":"o1","score":15.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"o3 & o4-mini","slug":"o3-o4-mini","url":"https://handyai.news/modeldrop/o3-o4-mini","published":"2025-04-16"},"launch_score":15.2,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":94},{"model":"Gemma 4","score":14.7,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":14.7,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":95},{"model":"Gemini 2.5 Pro Preview","score":14.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"first_cited","drop_url":null,"rank":96},{"model":"Grok Code Fast 1","score":14.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"GPT-5-Codex","slug":"gpt-5-codex","url":"https://handyai.news/modeldrop/gpt-5-codex","published":"2025-09-15"},"launch_score":14.1,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-08-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-code-fast-1","rank":97},{"model":"DeepSeek V3.1","score":13.5,"own_drop":true,"independent":true,"setting":"","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":13.5,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-08-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v31","rank":98},{"model":"Gemini 2.5 Flash","score":13.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":11.7,"reports":4,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-flash","rank":99},{"model":"DeepSeek R1 0528","score":13.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":13.1,"reports":6,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1-0528","rank":100},{"model":"Nemotron 3.5 Lightning","score":12.9,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":12.9,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-lightning","rank":101},{"model":"GLM-4.5","score":12.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":12.8,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-07-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-45","rank":102},{"model":"Nemotron 3 Super","score":12.8,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":12.8,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-super","rank":103},{"model":"GPT-4.1","score":12.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":12.7,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":104},{"model":"Kimi K2","score":12.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":12.7,"reports":3,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-07-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2","rank":105},{"model":"Qwen3-235B-A22B-Thinking-2507","score":12.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"first_cited","drop_url":null,"rank":106},{"model":"o1-pro","score":12.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"first_cited","drop_url":null,"rank":107},{"model":"Qwen3-Coder","score":11.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok Code Fast 1","slug":"grok-code-fast-1","url":"https://handyai.news/modeldrop/grok-code-fast-1","published":"2025-08-28"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-08-28","dated_by":"first_cited","drop_url":null,"rank":108},{"model":"gpt-oss-120b","score":11.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":11.6,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":109},{"model":"DeepSeek R1","score":11.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":11.4,"reports":9,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":110},{"model":"o3-mini","score":11,"own_drop":true,"independent":true,"setting":"","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":11,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":111},{"model":"Grok 3","score":10.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 2.5 Pro","slug":"gemini-25-pro","url":"https://handyai.news/modeldrop/gemini-25-pro","published":"2025-03-25"},"launch_score":10.4,"reports":4,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-3","rank":112},{"model":"GPT-4.1 mini","score":10.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-05-07","dated_by":"first_cited","drop_url":null,"rank":113},{"model":"Llama 4 Maverick","score":10,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":10,"reports":5,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":114},{"model":"o1-mini","score":9.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-20","dated_by":"first_cited","drop_url":null,"rank":115},{"model":"Qwen3-30B-A3B-Thinking-2507","score":9.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"first_cited","drop_url":null,"rank":116},{"model":"DeepSeek V3-0324","score":9.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"first_cited","drop_url":null,"rank":117},{"model":"DeepSeek V3 0324","score":9.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":null,"reports":3,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-14","dated_by":"first_cited","drop_url":null,"rank":118},{"model":"GPT-4.5","score":9.6,"own_drop":true,"independent":true,"setting":"","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":9.6,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":119},{"model":"Qwen3","score":9.5,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":9.5,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-04-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3","rank":120},{"model":"Gemini 2.0 Flash Thinking","score":9.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"first_cited","drop_url":null,"rank":121},{"model":"Mistral Medium 3","score":9,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":9,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2025-05-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-medium-3","rank":122},{"model":"gpt-oss-20b","score":9,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":123},{"model":"Gemini 2.0 Flash","score":8.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":8.2,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":124},{"model":"Claude 3.5 Haiku","score":8.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"first_cited","drop_url":null,"rank":125},{"model":"Nemotron 3 Nano","score":8.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":8.9,"reports":2,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":126},{"model":"Gemini 2.0 Pro","score":8.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 3","slug":"grok-3","url":"https://handyai.news/modeldrop/grok-3","published":"2025-02-17"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-02-17","dated_by":"first_cited","drop_url":null,"rank":127},{"model":"DeepSeek V3","score":8.5,"own_drop":true,"independent":true,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":8.5,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v3","rank":128},{"model":"GPT-4o","score":8.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":null,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":129},{"model":"Qwen3 235B","score":8.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"first_cited","drop_url":null,"rank":130},{"model":"Claude 3.5 Sonnet","score":7.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":131},{"model":"Gemini 1.5 Pro","score":7.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"o1","slug":"o1","url":"https://handyai.news/modeldrop/o1","published":"2024-12-05"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":132},{"model":"Qwen2.5 72B","score":7.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"first_cited","drop_url":null,"rank":133},{"model":"Llama 3.1 Nemotron Ultra","score":7.5,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Llama 3.1 Nemotron Ultra","slug":"llama-31-nemotron-ultra","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","published":"2025-04-07"},"launch_score":7.5,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","rank":134},{"model":"Nemotron Nano 2","score":7.4,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":7.4,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-nano-2","rank":135},{"model":"Llama 3.1 405B","score":7.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Llama 3.1 Nemotron Ultra","slug":"llama-31-nemotron-ultra","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","published":"2025-04-07"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"first_cited","drop_url":null,"rank":136},{"model":"GPT-4o mini","score":6.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"first_cited","drop_url":null,"rank":137},{"model":"Qwen3-8B","score":5.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"first_cited","drop_url":null,"rank":138}]},{"key":"gpqa-diamond","name":"GPQA Diamond","about":null,"type":{"key":"reasoning","label":"Reasoning"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#gpqa-diamond","api_url":"https://handyai.news/api/benchmarks/gpqa-diamond","image_url":"https://handyai.news/api/benchmarks/gpqa-diamond.png","leader":{"model":"GPT-6 Astra","score":96.1,"maker":"OpenAI"},"model_count":107,"drop_count":59,"first_date":"2024-12-05","last_date":"2026-10-28","frontier_gain":18.1,"frontier":[{"date":"2024-12-05","model":"o1","score":78,"maker":"openai"},{"date":"2025-02-17","model":"Grok 3","score":80.2,"maker":"xai"},{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":86.4,"maker":"google"},{"date":"2025-07-09","model":"Grok 4","score":87.5,"maker":"xai"},{"date":"2025-11-12","model":"GPT-5.1","score":88.1,"maker":"openai"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":91.9,"maker":"google"},{"date":"2025-12-11","model":"GPT-5.2 Pro","score":93.2,"maker":"openai"},{"date":"2026-02-19","model":"Gemini 3.1 Pro","score":94.3,"maker":"google"},{"date":"2026-04-07","model":"Claude Mythos Preview","score":94.5,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":96.1,"maker":"openai"}],"models":[{"model":"GPT-6 Astra","score":96.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":1},{"model":"Claude Mythos Preview","score":94.5,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Claude Mythos Preview","slug":"claude-mythos-preview","url":"https://handyai.news/modeldrop/claude-mythos-preview","published":"2026-04-07"},"launch_score":94.5,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":2},{"model":"Gemini 3.1 Pro","score":94.3,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":94.3,"reports":4,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":3},{"model":"Kimi K3.1 Pro","score":94.2,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":94.2,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":4},{"model":"GPT-5.6 Sol","score":94.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":5},{"model":"Claude Opus 5.5","score":93.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":6},{"model":"Fable 5.1","score":93.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":7},{"model":"Qwen3.8-2.4T-A95B","score":93.5,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":8},{"model":"Kimi K3","score":93.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":9},{"model":"GPT-5.2 Pro","score":93.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":10},{"model":"GPT-5.4","score":92.8,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":92.8,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":11},{"model":"Qwen3.8-Flash-Next","score":92.68,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"first_cited","drop_url":null,"rank":12},{"model":"GPT-5.3-Codex","score":92.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-53-codex","rank":13},{"model":"Qwen3.8-Max","score":92.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":92.6,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":14},{"model":"Fable 5","score":92.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":15},{"model":"GPT-5.2","score":92.4,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":92.4,"reports":6,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":16},{"model":"DeepSeek V4.1 Flash 1017","score":92.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":17},{"model":"Claude Opus 4.8","score":92,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-48","rank":18},{"model":"Gemini 3 Pro","score":91.9,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":91.9,"reports":8,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":19},{"model":"GLM-5.3","score":91.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":20},{"model":"Claude Opus 4.6","score":91.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":91.3,"reports":6,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":21},{"model":"Kimi K2.6","score":91.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-04-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k26","rank":22},{"model":"Gemini 3 Flash","score":90.4,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":90.4,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-flash","rank":23},{"model":"Particle-1","score":90.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Particle, \"Introducing Particle-1\"","slug":"particle-1","url":"https://particle.ai/blog/introducing-particle-1","published":"2026-10-28"},"launch_score":90.4,"reports":1,"maker":{"key":"particle","name":"Particle","logo_url":"https://handyai.news/ui/makers/particle.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":null,"rank":24},{"model":"Qwen3.7-Plus","score":90.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":25},{"model":"Claude Sonnet 4.6","score":89.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":89.9,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-46","rank":26},{"model":"Nemotron 3.5 Super VL","score":89.52,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":89.52,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":27},{"model":"Muse Spark","score":89.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":89.5,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-04-08","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark","rank":28},{"model":"GLM-5.2","score":89.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"first_cited","drop_url":null,"rank":29},{"model":"Nanbeige4.3-33B","score":89.5,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":89.5,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":30},{"model":"Qwen3.8","score":89.2,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":89.2,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":31},{"model":"Muse Spark Open","score":89,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":89,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":32},{"model":"GPT-5.1","score":88.1,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Claude Opus 4.5","slug":"claude-opus-45","url":"https://handyai.news/modeldrop/claude-opus-45","published":"2025-11-24"},"launch_score":88.1,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":33},{"model":"GPT-5.1 Thinking","score":88.1,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.2","slug":"gpt-52","url":"https://handyai.news/modeldrop/gpt-52","published":"2025-12-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"first_cited","drop_url":null,"rank":34},{"model":"Qwen3.6-27B","score":87.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":35},{"model":"Kimi K2.5","score":87.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/modeldrop/glm-5","published":"2026-02-11"},"launch_score":87.6,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":36},{"model":"Grok 4","score":87.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":87.5,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":37},{"model":"Inkling","score":87.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":87.2,"reports":1,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":38},{"model":"Qwen3.5 397B","score":87.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"first_cited","drop_url":null,"rank":39},{"model":"Claude Opus 4.5","score":87,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/modeldrop/glm-5","published":"2026-02-11"},"launch_score":87,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":40},{"model":"Nemotron 3 Ultra","score":86.99,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":87,"reports":3,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":41},{"model":"Qwen3.5-122B-A10B","score":86.74,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"first_cited","drop_url":null,"rank":42},{"model":"Gemini 2.5 Pro","score":86.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":84,"reports":8,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":43},{"model":"GLM-5.1","score":86.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":86.2,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-03-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-51","rank":44},{"model":"GLM-5","score":86,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/modeldrop/glm-5","published":"2026-02-11"},"launch_score":86,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-02-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-5","rank":45},{"model":"GPT-5","score":85.7,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":85.7,"reports":6,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":46},{"model":"Grok 4 Fast","score":85.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":85.7,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-09-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4-fast","rank":47},{"model":"MAI-Code-1-Flash","score":84.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":84.6,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-06-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-code-1-flash","rank":48},{"model":"Kimi K2 Thinking","score":84.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-11-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2-thinking","rank":49},{"model":"Grok 4.1 Fast","score":84.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"first_cited","drop_url":null,"rank":50},{"model":"Gemma 4","score":84.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":84.3,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":51},{"model":"MAI-Thinking-1","score":84.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":84.2,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-thinking-1","rank":52},{"model":"o3-pro","score":84,"own_drop":true,"independent":false,"setting":"","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":84,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-06-10","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-pro","rank":53},{"model":"Clover-35B-A3B-Preview","score":83.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DiffuSpace, \"Clover-35B-A3B-Preview\"","slug":"clover-35b-a3b-preview","url":"https://diffuspace.ai/blog/clover-35b/","published":"2026-10-28"},"launch_score":83.8,"reports":1,"maker":{"key":"diffuspace","name":"DiffuSpace","logo_url":"https://handyai.news/ui/makers/diffuspace.svg"},"open_weights":true,"released":"2026-10-28","dated_by":"own_drop","drop_url":null,"rank":54},{"model":"Muse Glimmer-30B","score":83.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":55},{"model":"Claude Sonnet 4.5","score":83.4,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":83.4,"reports":6,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":56},{"model":"Qwen3.6 35B A3B","score":83.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"first_cited","drop_url":null,"rank":57},{"model":"o3","score":83.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":83.3,"reports":8,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":58},{"model":"Gemini 2.5 Pro Preview","score":83,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"first_cited","drop_url":null,"rank":59},{"model":"DeepSeek V3.2","score":82.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":82.4,"reports":3,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-12-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v32","rank":60},{"model":"GPT-5 mini","score":82.3,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-28","dated_by":"first_cited","drop_url":null,"rank":61},{"model":"o4-mini","score":81.4,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":62},{"model":"DeepSeek R1 0528","score":81,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":81,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1-0528","rank":63},{"model":"Claude Opus 4.1","score":81,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":80.9,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":64},{"model":"Grok 3","score":80.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Pro","slug":"gemini-25-pro","url":"https://handyai.news/modeldrop/gemini-25-pro","published":"2025-03-25"},"launch_score":80.2,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-3","rank":65},{"model":"Grok 3 Beta","score":80.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":3,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"first_cited","drop_url":null,"rank":66},{"model":"gpt-oss-120b","score":80.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":80.1,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":67},{"model":"DeepSeek V3.1","score":80.1,"own_drop":true,"independent":false,"setting":"thinking","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":80.1,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-08-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v31","rank":68},{"model":"Gemma 4 26B A4B","score":79.61,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":69},{"model":"Claude Opus 4","score":79.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":79.6,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":70},{"model":"Nemotron 3 Super","score":79.23,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":79.23,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-super","rank":71},{"model":"o1-pro","score":79,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"first_cited","drop_url":null,"rank":72},{"model":"Grok 3 Mini","score":79,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-09-19","dated_by":"first_cited","drop_url":null,"rank":73},{"model":"Gemini 2.5 Flash","score":78.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":78.3,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-flash","rank":74},{"model":"Claude 3.7 Sonnet","score":78.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":78.2,"reports":5,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":75},{"model":"o1","score":78,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":75.7,"reports":7,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":76},{"model":"o3-mini","score":77,"own_drop":false,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":79.7,"reports":6,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":77},{"model":"Claude Sonnet 4","score":76.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":78},{"model":"Llama 3.1 Nemotron Ultra","score":76,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Llama 3.1 Nemotron Ultra","slug":"llama-31-nemotron-ultra","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","published":"2025-04-07"},"launch_score":76,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","rank":79},{"model":"Nemotron 3.5 Lightning","score":75.44,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":75.44,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-lightning","rank":80},{"model":"Kimi K2","score":75.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":75.1,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-07-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2","rank":81},{"model":"Nemotron 3 Nano","score":74.05,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":82},{"model":"o1-preview","score":73.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o1","slug":"o1","url":"https://handyai.news/modeldrop/o1","published":"2024-12-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":83},{"model":"Claude Haiku 4.5","score":73.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":73,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":84},{"model":"DeepSeek R1","score":71.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":71.5,"reports":8,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":85},{"model":"gpt-oss-20b","score":71.46,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":86},{"model":"Qwen3","score":71.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":71.1,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-04-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3","rank":87},{"model":"Qwen3 235B","score":71.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"first_cited","drop_url":null,"rank":88},{"model":"GPT-4.5","score":69.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":89},{"model":"DeepSeek V3-0324","score":68.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"first_cited","drop_url":null,"rank":90},{"model":"DeepSeek V3 0324","score":68.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-14","dated_by":"first_cited","drop_url":null,"rank":91},{"model":"GPT-4.1","score":66.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":66.3,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":92},{"model":"Claude 3.5 Sonnet","score":65,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude 3.7 Sonnet","slug":"claude-37-sonnet","url":"https://handyai.news/modeldrop/claude-37-sonnet","published":"2025-02-24"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":93},{"model":"Nemotron Nano 2","score":64,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":64,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-nano-2","rank":94},{"model":"LFM3-1.7B","score":63.37,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Liquid AI, \"Introducing LFM3\"","slug":"lfm3-17b","url":"https://www.liquid.ai/blog/lfm3","published":"2026-10-27"},"launch_score":63.37,"reports":1,"maker":{"key":"liquid","name":"Liquid AI","logo_url":"https://handyai.news/ui/makers/liquid.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":95},{"model":"Llama 4 Maverick","score":61.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":69.8,"reports":4,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":96},{"model":"Gemini 2.0 Flash","score":60.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":62.1,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":97},{"model":"o1-mini","score":60,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-20","dated_by":"first_cited","drop_url":null,"rank":98},{"model":"Qwen3-8B","score":59.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"first_cited","drop_url":null,"rank":99},{"model":"Gemini 1.5 Pro","score":59.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":100},{"model":"DeepSeek V3","score":59.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":59.1,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v3","rank":101},{"model":"Llama 4 Scout","score":57.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":57.2,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":102},{"model":"Mistral Medium 3","score":57.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":57.1,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2025-05-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-medium-3","rank":103},{"model":"GPT-4o","score":52.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":104},{"model":"Gemini 1.5 Flash","score":51,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"first_cited","drop_url":null,"rank":105},{"model":"Llama 3.1 405B","score":43.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Llama 3.1 Nemotron Ultra","slug":"llama-31-nemotron-ultra","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","published":"2025-04-07"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"first_cited","drop_url":null,"rank":106},{"model":"Gemma 3 27B","score":42.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"first_cited","drop_url":null,"rank":107}]},{"key":"swe-bench-verified","name":"SWE-bench Verified","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#swe-bench-verified","api_url":"https://handyai.news/api/benchmarks/swe-bench-verified","image_url":"https://handyai.news/api/benchmarks/swe-bench-verified.png","leader":{"model":"Claude Mythos Preview","score":93.9,"maker":"Anthropic"},"model_count":75,"drop_count":48,"first_date":"2024-12-05","last_date":"2026-10-28","frontier_gain":44.9,"frontier":[{"date":"2024-12-05","model":"Claude 3.5 Sonnet","score":49,"maker":"anthropic"},{"date":"2025-01-20","model":"DeepSeek R1","score":49.2,"maker":"deepseek"},{"date":"2025-01-31","model":"o3-mini","score":49.3,"maker":"openai"},{"date":"2025-02-24","model":"Claude 3.7 Sonnet","score":62.3,"maker":"anthropic"},{"date":"2025-04-16","model":"o3","score":69.1,"maker":"openai"},{"date":"2025-05-22","model":"Claude Sonnet 4","score":72.7,"maker":"anthropic"},{"date":"2025-08-05","model":"Claude Opus 4.1","score":74.5,"maker":"anthropic"},{"date":"2025-08-07","model":"GPT-5","score":74.9,"maker":"openai"},{"date":"2025-09-29","model":"Claude Sonnet 4.5","score":77.2,"maker":"anthropic"},{"date":"2025-11-19","model":"GPT-5.1-Codex-Max","score":77.9,"maker":"openai"},{"date":"2025-11-24","model":"Claude Opus 4.5","score":80.9,"maker":"anthropic"},{"date":"2026-04-07","model":"Claude Mythos Preview","score":93.9,"maker":"anthropic"}],"models":[{"model":"Claude Mythos Preview","score":93.9,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Claude Mythos Preview","slug":"claude-mythos-preview","url":"https://handyai.news/modeldrop/claude-mythos-preview","published":"2026-04-07"},"launch_score":93.9,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":1},{"model":"GPT-6 Astra","score":86,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mirendil, \"Introducing Fern\"","slug":"fern-01","url":"https://mirendil.com/news/introducing-fern/","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":2},{"model":"Claude Opus 4.5","score":80.9,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":80.9,"reports":5,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":3},{"model":"Claude Opus 4.6","score":80.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":80.8,"reports":6,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":4},{"model":"Gemini 3.1 Pro","score":80.6,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":80.6,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":5},{"model":"DeepSeek V4 Pro","score":80.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"first_cited","drop_url":null,"rank":6},{"model":"Kimi K2.6","score":80.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-04-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k26","rank":7},{"model":"GPT-5.2","score":80,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3.1 Pro","slug":"gemini-31-pro","url":"https://handyai.news/modeldrop/gemini-31-pro","published":"2026-02-19"},"launch_score":80,"reports":8,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":8},{"model":"Claude Sonnet 4.6","score":79.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":79.6,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-46","rank":9},{"model":"GLM-5.3-Flash","score":79.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"first_cited","drop_url":null,"rank":10},{"model":"Gemini 3 Flash","score":78,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":78,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-flash","rank":11},{"model":"GPT-5.1-Codex-Max","score":77.9,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Claude Opus 4.5","slug":"claude-opus-45","url":"https://handyai.news/modeldrop/claude-opus-45","published":"2025-11-24"},"launch_score":77.9,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51-codex-max","rank":12},{"model":"GLM-5","score":77.8,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/modeldrop/glm-5","published":"2026-02-11"},"launch_score":77.8,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-02-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-5","rank":13},{"model":"Inkling","score":77.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":77.6,"reports":1,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":14},{"model":"Muse Spark","score":77.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":77.4,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-04-08","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark","rank":15},{"model":"Claude Sonnet 4.5","score":77.2,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":77.2,"reports":9,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":16},{"model":"Kimi K2.5","score":76.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":76.8,"reports":3,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":17},{"model":"Qwen3.5","score":76.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":76.4,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-02-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen35","rank":18},{"model":"GPT-5.1","score":76.3,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Claude Opus 4.5","slug":"claude-opus-45","url":"https://handyai.news/modeldrop/claude-opus-45","published":"2025-11-24"},"launch_score":76.3,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":19},{"model":"GPT-5.1 Thinking","score":76.3,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.2","slug":"gpt-52","url":"https://handyai.news/modeldrop/gpt-52","published":"2025-12-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"first_cited","drop_url":null,"rank":20},{"model":"Gemini 3 Pro","score":76.2,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":76.2,"reports":7,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":21},{"model":"GPT-5","score":74.9,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":74.9,"reports":8,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":22},{"model":"Claude Opus 4.1","score":74.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":74.5,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":23},{"model":"GPT-5-Codex","score":74.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":74.5,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-09-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5-codex","rank":24},{"model":"Nemotron 3 Ultra","score":73.93,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":2,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":25},{"model":"GPT-5.1-Codex","score":73.7,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.1-Codex-Max","slug":"gpt-51-codex-max","url":"https://handyai.news/modeldrop/gpt-51-codex-max","published":"2025-11-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-19","dated_by":"first_cited","drop_url":null,"rank":26},{"model":"Clover-35B-A3B-Preview","score":73.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DiffuSpace, \"Clover-35B-A3B-Preview\"","slug":"clover-35b-a3b-preview","url":"https://diffuspace.ai/blog/clover-35b/","published":"2026-10-28"},"launch_score":73.6,"reports":1,"maker":{"key":"diffuspace","name":"DiffuSpace","logo_url":"https://handyai.news/ui/makers/diffuspace.svg"},"open_weights":true,"released":"2026-10-28","dated_by":"own_drop","drop_url":null,"rank":27},{"model":"MAI-Thinking-1","score":73.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":73.5,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-thinking-1","rank":28},{"model":"Fern 0.1","score":73.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mirendil, \"Introducing Fern\"","slug":"fern-01","url":"https://mirendil.com/news/introducing-fern/","published":"2026-10-26"},"launch_score":73.4,"reports":1,"maker":{"key":"mirendil","name":"Mirendil","logo_url":"https://handyai.news/ui/makers/mirendil.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":null,"rank":29},{"model":"DeepSeek V3.2","score":73.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":73.1,"reports":4,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-12-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v32","rank":30},{"model":"Claude Sonnet 4","score":72.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":null,"reports":6,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":31},{"model":"Claude Opus 4","score":72.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":72.5,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":32},{"model":"Qwen3.5-122B-A10B","score":71.87,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"first_cited","drop_url":null,"rank":33},{"model":"MAI-Code-1-Flash","score":71.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":71.6,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-06-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-code-1-flash","rank":34},{"model":"Kimi K2 Thinking","score":71.3,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":71.3,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-11-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2-thinking","rank":35},{"model":"Nemotron 3.5 Super VL","score":71.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":71.1,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":36},{"model":"Grok Code Fast 1","score":70.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok Code Fast 1","slug":"grok-code-fast-1","url":"https://handyai.news/modeldrop/grok-code-fast-1","published":"2025-08-28"},"launch_score":70.8,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-08-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-code-fast-1","rank":37},{"model":"Qwen3.6 35B A3B","score":70.12,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"first_cited","drop_url":null,"rank":38},{"model":"Qwen3-Max","score":69.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":69.6,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2025-09-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3-max","rank":39},{"model":"MiniMax-M2","score":69.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":69.4,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m2","rank":40},{"model":"Kimi K2 0905","score":69.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-09-24","dated_by":"first_cited","drop_url":null,"rank":41},{"model":"o3","score":69.1,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":69.1,"reports":6,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":42},{"model":"o4-mini","score":68.1,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":43},{"model":"GLM-4.6","score":68,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":44},{"model":"DeepSeek V3.2-Exp","score":67.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":45},{"model":"Claude Haiku 4.5","score":66.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":73.3,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":46},{"model":"DeepSeek V3.1","score":66,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":66,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-08-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v31","rank":47},{"model":"Kimi K2","score":65.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":65.8,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-07-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2","rank":48},{"model":"GLM-4.5","score":64.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":64.2,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-07-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-45","rank":49},{"model":"Gemini 2.5 Pro Preview","score":63.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"first_cited","drop_url":null,"rank":50},{"model":"Claude 3.7 Sonnet","score":62.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":62.3,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":51},{"model":"Nemotron 3 Super","score":60.47,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":60.47,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-super","rank":52},{"model":"Gemini 2.5 Flash","score":60.4,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-flash","rank":53},{"model":"Gemini 2.5 Pro","score":59.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":63.8,"reports":5,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":54},{"model":"Gemma 4 26B A4B","score":57.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":55},{"model":"GPT-4.1","score":54.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":54.6,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":56},{"model":"gpt-oss-20b","score":52.44,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":57},{"model":"Qwen3-235B-A22B-Instruct-2507","score":52.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-09-24","dated_by":"first_cited","drop_url":null,"rank":58},{"model":"Nemotron 3.5 Lightning","score":51.56,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":51.56,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-lightning","rank":59},{"model":"Grok 4.1 Fast","score":50.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"first_cited","drop_url":null,"rank":60},{"model":"o3-mini","score":49.3,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":49.3,"reports":7,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":61},{"model":"DeepSeek R1","score":49.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":49.2,"reports":4,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":62},{"model":"Claude 3.5 Sonnet","score":49,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude 3.7 Sonnet","slug":"claude-37-sonnet","url":"https://handyai.news/modeldrop/claude-37-sonnet","published":"2025-02-24"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":63},{"model":"o1","score":48.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3 & o4-mini","slug":"o3-o4-mini","url":"https://handyai.news/modeldrop/o3-o4-mini","published":"2025-04-16"},"launch_score":48.9,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":64},{"model":"DeepSeek V3 0324","score":45.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-14","dated_by":"first_cited","drop_url":null,"rank":65},{"model":"DeepSeek R1 0528","score":44.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":57.6,"reports":3,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1-0528","rank":66},{"model":"LFM3-1.7B","score":43.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Liquid AI, \"Introducing LFM3\"","slug":"lfm3-17b","url":"https://www.liquid.ai/blog/lfm3","published":"2026-10-27"},"launch_score":43.4,"reports":1,"maker":{"key":"liquid","name":"Liquid AI","logo_url":"https://handyai.news/ui/makers/liquid.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":67},{"model":"DeepSeek V3","score":42,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":42,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v3","rank":68},{"model":"gpt-oss-120b","score":41.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":62.4,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":69},{"model":"o1-mini","score":41.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1","slug":"deepseek-r1","url":"https://handyai.news/modeldrop/deepseek-r1","published":"2025-01-20"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-20","dated_by":"first_cited","drop_url":null,"rank":70},{"model":"o1-preview","score":41.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":71},{"model":"GPT-4.5","score":38,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":38,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":72},{"model":"Nemotron 3 Nano","score":34.08,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":73},{"model":"GPT-4o","score":33.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":74},{"model":"Llama 3.1 405B","score":24.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"first_cited","drop_url":null,"rank":75}]},{"key":"humanitys-last-exam-no-tools","name":"Humanity's Last Exam (no tools)","about":"Expert-level questions, no tools.","type":{"key":"reasoning","label":"Reasoning"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#humanitys-last-exam-no-tools","api_url":"https://handyai.news/api/benchmarks/humanitys-last-exam-no-tools","image_url":"https://handyai.news/api/benchmarks/humanitys-last-exam-no-tools.png","leader":{"model":"Claude Opus 5.5","score":64.4,"maker":"Anthropic"},"model_count":77,"drop_count":40,"first_date":"2024-12-11","last_date":"2026-10-27","frontier_gain":59.3,"frontier":[{"date":"2024-12-11","model":"Gemini 2.0 Flash","score":5.1,"maker":"google"},{"date":"2025-01-20","model":"DeepSeek R1","score":8.5,"maker":"deepseek"},{"date":"2025-01-31","model":"o3-mini","score":13.4,"maker":"openai"},{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":21.6,"maker":"google"},{"date":"2025-07-09","model":"Grok 4","score":25.4,"maker":"xai"},{"date":"2025-08-07","model":"GPT-5","score":26.3,"maker":"openai"},{"date":"2025-11-12","model":"GPT-5.1","score":26.5,"maker":"openai"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":37.5,"maker":"google"},{"date":"2026-02-05","model":"Claude Opus 4.6","score":40,"maker":"anthropic"},{"date":"2026-02-19","model":"Gemini 3.1 Pro","score":44.4,"maker":"google"},{"date":"2026-04-07","model":"Claude Mythos Preview","score":56.8,"maker":"anthropic"},{"date":"2026-06-09","model":"Mythos 5","score":59,"maker":"anthropic"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":64.4,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":64.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Opus 5.5","slug":"claude-opus-55","url":"https://handyai.news/modeldrop/claude-opus-55","published":"2026-09-22"},"launch_score":64.4,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"Mythos 5","score":59,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":59,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":2},{"model":"Fable 5","score":57.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5.1","slug":"fable-51","url":"https://handyai.news/modeldrop/fable-51","published":"2026-09-01"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":3},{"model":"Fable 5.1","score":57.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":60.9,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":4},{"model":"Claude Sonnet 5.5","score":56.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":5},{"model":"Claude Mythos Preview","score":56.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":56.8,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":6},{"model":"Claude Opus 5","score":56.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":7},{"model":"GPT-6 Astra","score":53.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":8},{"model":"Kimi K3.1 Pro","score":51.1,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":51.1,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":9},{"model":"Claude Haiku 5.5","score":45.9,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":45.9,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":10},{"model":"Claude Opus 4.8","score":45.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-48","rank":11},{"model":"GPT-5.6 Sol","score":44.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":12},{"model":"Gemini 3.1 Pro","score":44.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":44.4,"reports":5,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":13},{"model":"GPT-5.4","score":43.9,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":39.8,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":14},{"model":"Qwen3.8-Max","score":43.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":43.6,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":15},{"model":"Qwen3.8","score":43.6,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":43.6,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":16},{"model":"Kimi K3","score":43.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":17},{"model":"Muse Spark","score":42.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":42.8,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-04-08","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark","rank":18},{"model":"GPT-5.4 Pro","score":42.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":19},{"model":"DeepSeek V4.1 Flash 1017","score":42.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":42.6,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":20},{"model":"GPT-5.5","score":41.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-04-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-55","rank":21},{"model":"GLM-5.2","score":40.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"first_cited","drop_url":null,"rank":22},{"model":"Claude Opus 4.6","score":40,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":40,"reports":6,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":23},{"model":"Qwen3.8-Flash-Next","score":39.57,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"first_cited","drop_url":null,"rank":24},{"model":"Gemini 3 Pro","score":37.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":37.5,"reports":8,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":25},{"model":"DeepSeek V4.1 Flash","score":36.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":26},{"model":"GPT-5.2 Pro","score":36.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":27},{"model":"Kimi K2.6","score":35.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-04-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k26","rank":28},{"model":"GPT-5.2","score":34.5,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":34.5,"reports":7,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":29},{"model":"Gemini 3 Flash","score":33.7,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":33.7,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-flash","rank":30},{"model":"Claude Sonnet 4.6","score":33.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.1 Pro","slug":"gemini-31-pro","url":"https://handyai.news/modeldrop/gemini-31-pro","published":"2026-02-19"},"launch_score":33.2,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-46","rank":31},{"model":"Nemotron 3.5 Super VL","score":31.79,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":31.79,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":32},{"model":"Kimi K2.5","score":31.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":30.1,"reports":4,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":33},{"model":"Claude Opus 4.5","score":30.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":34},{"model":"GLM-5","score":30.5,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/modeldrop/glm-5","published":"2026-02-11"},"launch_score":30.5,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-02-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-5","rank":35},{"model":"Inkling","score":29.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":29.7,"reports":1,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":36},{"model":"Nemotron 3 Ultra","score":29.24,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":26.7,"reports":3,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":37},{"model":"Qwen3.5","score":28.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":28.7,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-02-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen35","rank":38},{"model":"Qwen3.5 397B","score":28.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"first_cited","drop_url":null,"rank":39},{"model":"GLM-5.1","score":27.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":31,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-03-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-51","rank":40},{"model":"GPT-5.1","score":26.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":41},{"model":"GPT-5","score":26.3,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":24.8,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":42},{"model":"GPT-5.1 Thinking","score":25.7,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.2","slug":"gpt-52","url":"https://handyai.news/modeldrop/gpt-52","published":"2025-12-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"first_cited","drop_url":null,"rank":43},{"model":"Grok 4","score":25.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":25.4,"reports":4,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":44},{"model":"DeepSeek V3.2","score":25.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":25.1,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-12-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v32","rank":45},{"model":"Qwen3.5-122B-A10B","score":24.24,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"first_cited","drop_url":null,"rank":46},{"model":"Kimi K2 Thinking","score":23.9,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":23.9,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-11-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2-thinking","rank":47},{"model":"Gemini 2.5 Pro","score":21.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":18.8,"reports":5,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":48},{"model":"o3","score":20.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":20.32,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":49},{"model":"Grok 4 Fast","score":20,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":20,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-09-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4-fast","rank":50},{"model":"Gemma 4","score":19.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":19.5,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":51},{"model":"Nemotron 3 Super","score":18.26,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":18.26,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-super","rank":52},{"model":"DeepSeek R1 0528","score":17.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":17.7,"reports":3,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1-0528","rank":53},{"model":"Claude Sonnet 4.5","score":17.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":null,"reports":5,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":54},{"model":"Grok 4.1 Fast","score":17.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"first_cited","drop_url":null,"rank":55},{"model":"GPT-5 mini","score":16.7,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-28","dated_by":"first_cited","drop_url":null,"rank":56},{"model":"DeepSeek V3.1","score":15.9,"own_drop":true,"independent":false,"setting":"thinking","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":15.9,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-08-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v31","rank":57},{"model":"gpt-oss-120b","score":14.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":58},{"model":"GLM-4.5","score":14.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":14.4,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-07-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-45","rank":59},{"model":"o4-mini","score":14.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":60},{"model":"o3-mini","score":13.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3 & o4-mini","slug":"o3-o4-mini","url":"https://handyai.news/modeldrop/o3-o4-mini","published":"2025-04-16"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":61},{"model":"Gemini 2.5 Flash","score":12.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":12.1,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-flash","rank":62},{"model":"Qwen3 235B","score":11.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"first_cited","drop_url":null,"rank":63},{"model":"gpt-oss-20b","score":10.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":64},{"model":"Nemotron 3 Nano","score":10.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":10.6,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":65},{"model":"Claude Haiku 4.5","score":10.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":66},{"model":"Qwen3-30B-A3B-Thinking-2507","score":9.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"first_cited","drop_url":null,"rank":67},{"model":"Claude 3.7 Sonnet","score":8.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":68},{"model":"Gemma 4 26B A4B","score":8.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":69},{"model":"DeepSeek R1","score":8.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":70},{"model":"o1-pro","score":8.12,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3 & o4-mini","slug":"o3-o4-mini","url":"https://handyai.news/modeldrop/o3-o4-mini","published":"2025-04-16"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"first_cited","drop_url":null,"rank":71},{"model":"LFM3-1.7B","score":7.17,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Liquid AI, \"Introducing LFM3\"","slug":"lfm3-17b","url":"https://www.liquid.ai/blog/lfm3","published":"2026-10-27"},"launch_score":7.17,"reports":1,"maker":{"key":"liquid","name":"Liquid AI","logo_url":"https://handyai.news/ui/makers/liquid.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":72},{"model":"Nemotron Nano 2","score":6.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":6.5,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-nano-2","rank":73},{"model":"GPT-4.5","score":6.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Pro","slug":"gemini-25-pro","url":"https://handyai.news/modeldrop/gemini-25-pro","published":"2025-03-25"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":74},{"model":"GPT-4.1","score":5.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":75},{"model":"Gemini 2.0 Flash","score":5.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":76},{"model":"Qwen3-8B","score":4.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"first_cited","drop_url":null,"rank":77}]},{"key":"livecodebench","name":"LiveCodeBench","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#livecodebench","api_url":"https://handyai.news/api/benchmarks/livecodebench","image_url":"https://handyai.news/api/benchmarks/livecodebench.png","leader":{"model":"DeepSeek V4","score":93.5,"maker":"DeepSeek"},"model_count":61,"drop_count":24,"first_date":"2024-12-05","last_date":"2026-10-27","frontier_gain":29.6,"frontier":[{"date":"2024-12-05","model":"o1","score":63.9,"maker":"openai"},{"date":"2025-01-31","model":"o3-mini","score":74.1,"maker":"openai"},{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":74.2,"maker":"google"},{"date":"2025-07-09","model":"Grok 4","score":79,"maker":"xai"},{"date":"2025-08-07","model":"GPT-5","score":84.5,"maker":"openai"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":90.7,"maker":"google"},{"date":"2026-04-24","model":"DeepSeek V4","score":93.5,"maker":"deepseek"}],"models":[{"model":"DeepSeek V4","score":93.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-04-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v4","rank":1},{"model":"Gemini 3 Pro","score":90.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":2},{"model":"Kimi K2.6","score":89.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-04-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k26","rank":3},{"model":"Nemotron 3 Ultra","score":89,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":89,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":4},{"model":"GPT-5.2","score":87.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":5},{"model":"MAI-Thinking-1","score":87.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":87.7,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-thinking-1","rank":6},{"model":"GLM-5.1","score":85.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-03-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-51","rank":7},{"model":"Kimi K2.5","score":85,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":8},{"model":"Claude Opus 4.5","score":84.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":9},{"model":"GPT-5","score":84.5,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":10},{"model":"Qwen3.5","score":83.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":83.6,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-02-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen35","rank":11},{"model":"DeepSeek V3.2","score":83.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":83.3,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-12-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v32","rank":12},{"model":"Kimi K2 Thinking","score":82.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":83.1,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-11-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2-thinking","rank":13},{"model":"Grok 4 Fast","score":80,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":80,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-09-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4-fast","rank":14},{"model":"Gemma 4","score":80,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":80,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":15},{"model":"Qwen3.5 397B","score":79.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"first_cited","drop_url":null,"rank":16},{"model":"Grok 4","score":79,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":79,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":17},{"model":"GPT-5 mini","score":77.4,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-28","dated_by":"first_cited","drop_url":null,"rank":18},{"model":"Gemma 4 26B A4B","score":77.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":19},{"model":"Gemini 2.5 Pro","score":74.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4","slug":"grok-4","url":"https://handyai.news/modeldrop/grok-4","published":"2025-07-09"},"launch_score":70.4,"reports":4,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":20},{"model":"o3-mini","score":74.1,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Gemini 2.5 Pro","slug":"gemini-25-pro","url":"https://handyai.news/modeldrop/gemini-25-pro","published":"2025-03-25"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":21},{"model":"DeepSeek V3.2-Exp","score":74.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":22},{"model":"LFM3-1.7B","score":73.46,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Liquid AI, \"Introducing LFM3\"","slug":"lfm3-17b","url":"https://www.liquid.ai/blog/lfm3","published":"2026-10-27"},"launch_score":73.46,"reports":1,"maker":{"key":"liquid","name":"Liquid AI","logo_url":"https://handyai.news/ui/makers/liquid.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":23},{"model":"DeepSeek R1 0528","score":73.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":73.3,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1-0528","rank":24},{"model":"o3","score":72,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4","slug":"grok-4","url":"https://handyai.news/modeldrop/grok-4","published":"2025-07-09"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":25},{"model":"Nemotron Nano 2","score":71.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":71.1,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-nano-2","rank":26},{"model":"Qwen3","score":70.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":70.7,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-04-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3","rank":27},{"model":"Grok 3","score":70.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Pro","slug":"gemini-25-pro","url":"https://handyai.news/modeldrop/gemini-25-pro","published":"2025-03-25"},"launch_score":70.6,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-3","rank":28},{"model":"Grok 3 Beta","score":70.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"first_cited","drop_url":null,"rank":29},{"model":"Qwen3-Max","score":69,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":69,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2025-09-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3-max","rank":30},{"model":"Nemotron 3 Nano","score":68.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":68.3,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":31},{"model":"Qwen3 235B","score":66.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"first_cited","drop_url":null,"rank":32},{"model":"Llama 3.1 Nemotron Ultra","score":66.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Llama 3.1 Nemotron Ultra","slug":"llama-31-nemotron-ultra","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","published":"2025-04-07"},"launch_score":66.3,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","rank":33},{"model":"Qwen3-30B-A3B-Thinking-2507","score":66,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"first_cited","drop_url":null,"rank":34},{"model":"Claude Sonnet 4.5","score":64,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":35},{"model":"o1","score":63.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":36},{"model":"DeepSeek R1","score":63.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":65.9,"reports":6,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":37},{"model":"Gemini 2.5 Flash","score":63.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":63.5,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-flash","rank":38},{"model":"gpt-oss-20b","score":61,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":39},{"model":"Qwen3-8B","score":59.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"first_cited","drop_url":null,"rank":40},{"model":"o1-mini","score":53.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1","slug":"deepseek-r1","url":"https://handyai.news/modeldrop/deepseek-r1","published":"2025-01-20"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-20","dated_by":"first_cited","drop_url":null,"rank":41},{"model":"Kimi K2","score":53.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":53.7,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-07-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2","rank":42},{"model":"DeepSeek V3.1","score":52.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-08-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v31","rank":43},{"model":"Qwen3-235B-A22B-Instruct-2507","score":51.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-09-24","dated_by":"first_cited","drop_url":null,"rank":44},{"model":"DeepSeek V3 0324","score":46.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-14","dated_by":"first_cited","drop_url":null,"rank":45},{"model":"Gemini 2.0 Flash Thinking","score":45.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 3","slug":"grok-3","url":"https://handyai.news/modeldrop/grok-3","published":"2025-02-17"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"first_cited","drop_url":null,"rank":46},{"model":"DeepSeek V3-0324","score":45.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"first_cited","drop_url":null,"rank":47},{"model":"GPT-4.1","score":44.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":48},{"model":"Claude Opus 4","score":44.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":49},{"model":"DeepSeek V3","score":40.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":40.5,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v3","rank":50},{"model":"Claude 3.5 Sonnet","score":36.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":51},{"model":"Claude 3.7 Sonnet","score":36,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":52},{"model":"Gemini 2.0 Flash","score":34.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":35.1,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":53},{"model":"Gemini 1.5 Pro","score":34.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":54},{"model":"Llama 4 Scout","score":32.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":32.8,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":55},{"model":"GPT-4o","score":31.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":56},{"model":"Mistral Medium 3","score":30.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":30.3,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2025-05-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-medium-3","rank":57},{"model":"Gemini 1.5 Flash","score":30,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"first_cited","drop_url":null,"rank":58},{"model":"Gemma 3 27B","score":29.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"first_cited","drop_url":null,"rank":59},{"model":"Llama 4 Maverick","score":28.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":43.4,"reports":3,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":60},{"model":"Llama 3.1 405B","score":28.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"first_cited","drop_url":null,"rank":61}]},{"key":"aime-2025","name":"AIME 2025","about":null,"type":{"key":"math","label":"Math"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#aime-2025","api_url":"https://handyai.news/api/benchmarks/aime-2025","image_url":"https://handyai.news/api/benchmarks/aime-2025.png","leader":{"model":"Claude Opus 4.6","score":99.8,"maker":"Anthropic"},"model_count":46,"drop_count":24,"first_date":"2024-12-05","last_date":"2026-08-12","frontier_gain":20.6,"frontier":[{"date":"2024-12-05","model":"o1","score":79.2,"maker":"openai"},{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":88,"maker":"google"},{"date":"2025-04-16","model":"o4-mini","score":99.5,"maker":"openai"},{"date":"2026-02-05","model":"Claude Opus 4.6","score":99.8,"maker":"anthropic"}],"models":[{"model":"Claude Opus 4.6","score":99.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":1},{"model":"o4-mini","score":99.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":2},{"model":"gpt-oss-120b","score":97.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":97.9,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":3},{"model":"MAI-Thinking-1","score":97,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":97,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-thinking-1","rank":4},{"model":"Claude Sonnet 4.6","score":95.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-46","rank":5},{"model":"Gemini 3 Pro","score":95,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":6},{"model":"GPT-5","score":94.6,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":94.6,"reports":7,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":7},{"model":"Kimi K2 Thinking","score":94.5,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":94.5,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-11-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2-thinking","rank":8},{"model":"GPT-5.1","score":94,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"GPT-5.1","slug":"gpt-51","url":"https://handyai.news/modeldrop/gpt-51","published":"2025-11-12"},"launch_score":94,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":9},{"model":"DeepSeek V3.2","score":93.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":93.1,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-12-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v32","rank":10},{"model":"Grok 4 Fast","score":92,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":92,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-09-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4-fast","rank":11},{"model":"Grok 4","score":91.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":91.7,"reports":3,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":12},{"model":"gpt-oss-20b","score":91.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":13},{"model":"GPT-5 mini","score":91.1,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-28","dated_by":"first_cited","drop_url":null,"rank":14},{"model":"DeepSeek V3.2-Exp","score":89.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":15},{"model":"Nemotron 3 Nano","score":89.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":89.1,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":16},{"model":"o3","score":88.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":88.9,"reports":7,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":17},{"model":"Gemini 2.5 Pro","score":88,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":86.7,"reports":6,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":18},{"model":"DeepSeek R1 0528","score":87.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":87.5,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1-0528","rank":19},{"model":"Claude Sonnet 4.5","score":87,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":87,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":20},{"model":"Qwen3-30B-A3B-Thinking-2507","score":85,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"first_cited","drop_url":null,"rank":21},{"model":"Gemini 2.5 Pro Preview","score":83,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"first_cited","drop_url":null,"rank":22},{"model":"Qwen3-Max","score":81.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":81.6,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2025-09-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3-max","rank":23},{"model":"Qwen3","score":81.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":81.5,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-04-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3","rank":24},{"model":"Qwen3 235B","score":81.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"first_cited","drop_url":null,"rank":25},{"model":"Claude Haiku 4.5","score":80.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":80.7,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":26},{"model":"o1","score":79.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":27},{"model":"Gemini 2.5 Flash","score":78,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":78,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-flash","rank":28},{"model":"Claude Opus 4.1","score":78,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":78,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":29},{"model":"Grok 3","score":77.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Pro","slug":"gemini-25-pro","url":"https://handyai.news/modeldrop/gemini-25-pro","published":"2025-03-25"},"launch_score":77.3,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-3","rank":30},{"model":"Grok 3 Beta","score":77.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"first_cited","drop_url":null,"rank":31},{"model":"o3-mini","score":74.8,"own_drop":false,"independent":false,"setting":"medium","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":32},{"model":"Llama 3.1 Nemotron Ultra","score":72.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Llama 3.1 Nemotron Ultra","slug":"llama-31-nemotron-ultra","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","published":"2025-04-07"},"launch_score":72.5,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","rank":33},{"model":"Nemotron Nano 2","score":72.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":72.1,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-nano-2","rank":34},{"model":"Claude Sonnet 4","score":70.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":35},{"model":"Qwen3-235B-A22B-Instruct-2507","score":70.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-09-24","dated_by":"first_cited","drop_url":null,"rank":36},{"model":"DeepSeek R1","score":70,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":null,"reports":6,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":37},{"model":"Qwen3-8B","score":69.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"first_cited","drop_url":null,"rank":38},{"model":"Claude 3.7 Sonnet","score":54.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":39},{"model":"Gemini 2.0 Flash Thinking","score":53.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 3","slug":"grok-3","url":"https://handyai.news/modeldrop/grok-3","published":"2025-02-17"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"first_cited","drop_url":null,"rank":40},{"model":"DeepSeek V3.1","score":49.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":88.4,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-08-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v31","rank":41},{"model":"Kimi K2","score":49.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":49.5,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-07-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2","rank":42},{"model":"DeepSeek V3 0324","score":46.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-14","dated_by":"first_cited","drop_url":null,"rank":43},{"model":"GPT-4.1","score":46.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":44},{"model":"Claude Opus 4","score":33.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":75.5,"reports":5,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":45},{"model":"Gemini 2.0 Flash","score":27.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":46}]},{"key":"deepswe-v1-1","name":"DeepSWE v1.1","about":"Fixing issues in real repositories.","type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#deepswe-v1-1","api_url":"https://handyai.news/api/benchmarks/deepswe-v1-1","image_url":"https://handyai.news/api/benchmarks/deepswe-v1-1.png","leader":{"model":"Muse Wonder Full","score":79,"maker":"Meta"},"model_count":45,"drop_count":24,"first_date":"2026-04-02","last_date":"2026-10-28","frontier_gain":76.3,"frontier":[{"date":"2026-04-02","model":"Gemma 4","score":2.7,"maker":"google"},{"date":"2026-04-23","model":"GPT-5.5","score":67,"maker":"openai"},{"date":"2026-06-09","model":"Fable 5","score":69.9,"maker":"anthropic"},{"date":"2026-06-26","model":"GPT-5.6 Sol","score":73,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":74,"maker":"anthropic"},{"date":"2026-09-02","model":"Muse Spark 1.3","score":75.4,"maker":"meta"},{"date":"2026-10-01","model":"Gemini 4 Argon","score":77.9,"maker":"google"},{"date":"2026-10-12","model":"Muse Wonder Full","score":79,"maker":"meta"}],"models":[{"model":"Muse Wonder Full","score":79,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":1},{"model":"Muse Wonder","score":78.8,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":78.8,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":2},{"model":"Gemini 4 Argon","score":77.9,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Argon","slug":"gemini-4-argon","url":"https://handyai.news/modeldrop/gemini-4-argon","published":"2026-10-01"},"launch_score":77.9,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":3},{"model":"DeepSeek V4.1 Flash 1017","score":77,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":77,"reports":3,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":4},{"model":"Kimi K3.1 Pro","score":75.5,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":75.5,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":5},{"model":"Muse Spark 1.3","score":75.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6 Astra","slug":"gpt-6-astra","url":"https://handyai.news/modeldrop/gpt-6-astra","published":"2026-09-03"},"launch_score":75.4,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-13","rank":6},{"model":"GPT-6.1 Sol","score":75.2,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"GPT-6.1 Sol","slug":"gpt-61-sol","url":"https://handyai.news/modeldrop/gpt-61-sol","published":"2026-09-29"},"launch_score":75.2,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":7},{"model":"Claude Opus 5.5","score":74.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":8},{"model":"DeepSeek V4.1 Flash","score":74.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":9},{"model":"GPT-6 Astra","score":74.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":74.1,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":10},{"model":"Claude Opus 5","score":74,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":8,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":11},{"model":"Gemini 3.8 Flash","score":73.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-6 Astra","slug":"gpt-6-astra","url":"https://handyai.news/modeldrop/gpt-6-astra","published":"2026-09-03"},"launch_score":73.7,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":12},{"model":"MiniMax M3.1 Flash","score":73.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":73.7,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":13},{"model":"GPT-5.6 Sol","score":73,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":72.7,"reports":12,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":14},{"model":"Particle-1","score":72.57,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Particle, \"Introducing Particle-1\"","slug":"particle-1","url":"https://particle.ai/blog/introducing-particle-1","published":"2026-10-28"},"launch_score":72.57,"reports":1,"maker":{"key":"particle","name":"Particle","logo_url":"https://handyai.news/ui/makers/particle.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":null,"rank":15},{"model":"MiMo-V2.6-Pro","score":71.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":71.9,"reports":1,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":16},{"model":"Grok 4.7","score":71,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":71,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":17},{"model":"Fable 5","score":69.9,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-6 Sol & GPT-6 Luna","slug":"gpt-6-sol-and-gpt-6-luna","url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","published":"2026-09-22"},"launch_score":null,"reports":6,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":18},{"model":"GPT-5.6 Terra","score":69.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":19},{"model":"GPT-6 Sol","score":68.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Sol","slug":"gpt-61-sol","url":"https://handyai.news/modeldrop/gpt-61-sol","published":"2026-09-29"},"launch_score":68.8,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":20},{"model":"Kimi K3","score":67.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":6,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":21},{"model":"Fable 5.1","score":67.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":22},{"model":"GPT-5.6 Luna","score":67,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":23},{"model":"GPT-5.5","score":67,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.6 Sol, Terra, & Luna","slug":"gpt-56-sol-terra-and-luna","url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","published":"2026-06-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-04-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-55","rank":24},{"model":"GLM-5.3","score":66.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":66.9,"reports":6,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":25},{"model":"GPT-6 Luna","score":66.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6 Sol & GPT-6 Luna","slug":"gpt-6-sol-and-gpt-6-luna","url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","published":"2026-09-22"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":26},{"model":"Nanbeige4.3-33B","score":65.5,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":65.5,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":27},{"model":"Gemini 3.7 Flash","score":65.3,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":65.3,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-08-13","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-37-flash","rank":28},{"model":"Grok 4.6","score":65.2,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":65.9,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":29},{"model":"Mistral Large 4","score":62,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":61.7,"reports":2,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":30},{"model":"Muse Spark Open","score":60,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":60,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":31},{"model":"Claude Opus 4.8","score":59,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-48","rank":32},{"model":"DeepSeek V4 Pro 0813","score":57,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"first_cited","drop_url":null,"rank":33},{"model":"Qwen3.8-2.4T-A95B","score":57,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":34},{"model":"Qwen3.8-Max","score":56.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":56.6,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":35},{"model":"Muse Spark 1.2","score":55,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":59.3,"reports":3,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-12","rank":36},{"model":"Grok 4.5","score":54,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":3,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":37},{"model":"Claude Sonnet 5","score":53.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":38},{"model":"Gemini 3.6 Flash","score":48.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":49,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":39},{"model":"Fern 0.1","score":44.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mirendil, \"Introducing Fern\"","slug":"fern-01","url":"https://mirendil.com/news/introducing-fern/","published":"2026-10-26"},"launch_score":44.2,"reports":1,"maker":{"key":"mirendil","name":"Mirendil","logo_url":"https://handyai.news/ui/makers/mirendil.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":null,"rank":40},{"model":"Qwen3.8","score":42.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":41},{"model":"Gemini 3.5 Flash","score":37,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-05-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-35-flash","rank":42},{"model":"Muse Glimmer-30B","score":4.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":43},{"model":"Nemotron 3 Ultra","score":3.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mirendil, \"Introducing Fern\"","slug":"fern-01","url":"https://mirendil.com/news/introducing-fern/","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":44},{"model":"Gemma 4","score":2.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":45}]},{"key":"terminal-bench-2-0","name":"Terminal-Bench 2.0","about":"Hard tasks in a real terminal (2025 edition).","type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#terminal-bench-2-0","api_url":"https://handyai.news/api/benchmarks/terminal-bench-2-0","image_url":"https://handyai.news/api/benchmarks/terminal-bench-2-0.png","leader":{"model":"Claude Mythos Preview","score":82,"maker":"Anthropic"},"model_count":33,"drop_count":22,"first_date":"2025-03-25","last_date":"2026-08-12","frontier_gain":49.4,"frontier":[{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":32.6,"maker":"google"},{"date":"2025-08-07","model":"GPT-5","score":35.2,"maker":"openai"},{"date":"2025-09-29","model":"Claude Sonnet 4.5","score":51,"maker":"anthropic"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":56.2,"maker":"google"},{"date":"2025-11-19","model":"GPT-5.1-Codex-Max","score":58.1,"maker":"openai"},{"date":"2025-12-11","model":"GPT-5.2","score":62.2,"maker":"openai"},{"date":"2025-12-18","model":"GPT-5.2-Codex","score":64,"maker":"openai"},{"date":"2026-02-05","model":"GPT-5.3-Codex","score":77.3,"maker":"openai"},{"date":"2026-04-07","model":"Claude Mythos Preview","score":82,"maker":"anthropic"}],"models":[{"model":"Claude Mythos Preview","score":82,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Claude Mythos Preview","slug":"claude-mythos-preview","url":"https://handyai.news/modeldrop/claude-mythos-preview","published":"2026-04-07"},"launch_score":82,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":1},{"model":"GPT-5.3-Codex","score":77.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":77.3,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-53-codex","rank":2},{"model":"GPT-5.4","score":75.1,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":75.1,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":3},{"model":"Gemini 3.1 Pro","score":68.5,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":68.5,"reports":4,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":4},{"model":"Claude Opus 4.6","score":65.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":65.4,"reports":7,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":5},{"model":"GPT-5.2-Codex","score":64,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.3-Codex","slug":"gpt-53-codex","url":"https://handyai.news/modeldrop/gpt-53-codex","published":"2026-02-05"},"launch_score":64,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52-codex","rank":6},{"model":"GLM-5.1","score":63.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":63.5,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-03-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-51","rank":7},{"model":"GPT-5.2","score":62.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":null,"reports":9,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":8},{"model":"Composer 2","score":61.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Composer 2","slug":"composer-2","url":"https://handyai.news/modeldrop/composer-2","published":"2026-03-19"},"launch_score":61.7,"reports":1,"maker":{"key":"cursor","name":"Cursor","logo_url":"https://handyai.news/ui/makers/cursor.svg"},"open_weights":false,"released":"2026-03-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/composer-2","rank":9},{"model":"Claude Sonnet 4.6","score":59.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":59.1,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-46","rank":10},{"model":"Muse Spark","score":59,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":59,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-04-08","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark","rank":11},{"model":"GPT-5.1-Codex-Max","score":58.1,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":58.1,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51-codex-max","rank":12},{"model":"Gemini 3 Pro","score":56.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":54.2,"reports":9,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":13},{"model":"GLM-5","score":56.2,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/modeldrop/glm-5","published":"2026-02-11"},"launch_score":56.2,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-02-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-5","rank":14},{"model":"MAI-Code-1-Flash","score":54.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":54.8,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-06-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-code-1-flash","rank":15},{"model":"GPT-5.1-Codex","score":52.8,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.1-Codex-Max","slug":"gpt-51-codex-max","url":"https://handyai.news/modeldrop/gpt-51-codex-max","published":"2025-11-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-19","dated_by":"first_cited","drop_url":null,"rank":16},{"model":"Qwen3.5","score":52.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":52.5,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-02-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen35","rank":17},{"model":"Claude Opus 4.5","score":52.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Composer 2","slug":"composer-2","url":"https://handyai.news/modeldrop/composer-2","published":"2026-03-19"},"launch_score":59.3,"reports":6,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":18},{"model":"Claude Sonnet 4.5","score":51,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":null,"reports":5,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":19},{"model":"Kimi K2.5","score":50.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":50.8,"reports":4,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":20},{"model":"GPT-5.1","score":47.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.5","slug":"claude-opus-45","url":"https://handyai.news/modeldrop/claude-opus-45","published":"2025-11-24"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":21},{"model":"Gemini 3 Flash","score":47.6,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":47.6,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-flash","rank":22},{"model":"DeepSeek V3.2","score":46.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":46.4,"reports":3,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-12-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v32","rank":23},{"model":"MAI-Thinking-1","score":46,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":46,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-thinking-1","rank":24},{"model":"Claude Haiku 4.5","score":41.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":25},{"model":"Composer","score":40,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Composer","slug":"composer","url":"https://handyai.news/modeldrop/composer","published":"2025-10-29"},"launch_score":40,"reports":1,"maker":{"key":"cursor","name":"Cursor","logo_url":"https://handyai.news/ui/makers/cursor.svg"},"open_weights":false,"released":"2025-10-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/composer","rank":26},{"model":"Qwen3.5-122B-A10B","score":37.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"first_cited","drop_url":null,"rank":27},{"model":"Kimi K2 Thinking","score":35.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-11-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2-thinking","rank":28},{"model":"GPT-5","score":35.2,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":29},{"model":"Gemini 2.5 Pro","score":32.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":30},{"model":"Nemotron 3 Super","score":31,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":31,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-super","rank":31},{"model":"gpt-oss-120b","score":18.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":32},{"model":"Gemini 2.5 Flash","score":16.9,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-flash","rank":33}]},{"key":"browsecomp","name":"BrowseComp","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#browsecomp","api_url":"https://handyai.news/api/benchmarks/browsecomp","image_url":"https://handyai.news/api/benchmarks/browsecomp.png","leader":{"model":"Kimi K3.1 Pro","score":92.1,"maker":"Moonshot AI"},"model_count":49,"drop_count":21,"first_date":"2024-12-05","last_date":"2026-11-02","frontier_gain":90.2,"frontier":[{"date":"2024-12-05","model":"GPT-4o","score":1.9,"maker":"openai"},{"date":"2025-04-16","model":"deep research","score":51.5,"maker":null},{"date":"2025-08-07","model":"ChatGPT agent","score":68.9,"maker":"openai"},{"date":"2025-12-11","model":"GPT-5.2 Pro","score":77.9,"maker":"openai"},{"date":"2026-02-05","model":"Claude Opus 4.6","score":84,"maker":"anthropic"},{"date":"2026-02-19","model":"Gemini 3.1 Pro","score":85.9,"maker":"google"},{"date":"2026-06-26","model":"GPT-5.6 Sol","score":90.4,"maker":"openai"},{"date":"2026-09-03","model":"GPT-6 Astra","score":91.5,"maker":"openai"},{"date":"2026-10-23","model":"Kimi K3.1 Pro","score":92.1,"maker":"moonshot"}],"models":[{"model":"Kimi K3.1 Pro","score":92.1,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":92.1,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":1},{"model":"LongCat-2.5","score":91.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Meituan LongCat-2.5 launch post","slug":"longcat-2-5","url":"https://x.com/Meituan_LongCat/status/2117264311714537633","published":"2026-11-02"},"launch_score":91.9,"reports":1,"maker":{"key":"meituan","name":"Meituan","logo_url":"https://handyai.news/ui/makers/meituan.svg"},"open_weights":true,"released":"2026-11-02","dated_by":"own_drop","drop_url":null,"rank":2},{"model":"GPT-6 Astra","score":91.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":3},{"model":"Claude Opus 5.5","score":90.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":4},{"model":"GPT-5.6 Sol","score":90.4,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"GPT-5.6 Sol, Terra, & Luna","slug":"gpt-56-sol-terra-and-luna","url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","published":"2026-06-26"},"launch_score":90.4,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":5},{"model":"GPT-5.6 Terra","score":87.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-5.6 Sol, Terra, & Luna","slug":"gpt-56-sol-terra-and-luna","url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","published":"2026-06-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":6},{"model":"Gemini 3.1 Pro","score":85.9,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 3.1 Pro","slug":"gemini-31-pro","url":"https://handyai.news/modeldrop/gemini-31-pro","published":"2026-02-19"},"launch_score":85.9,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":7},{"model":"GPT-5.5","score":84.4,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.6 Sol, Terra, & Luna","slug":"gpt-56-sol-terra-and-luna","url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","published":"2026-06-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-04-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-55","rank":8},{"model":"Claude Opus 4.6","score":84,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.1 Pro","slug":"gemini-31-pro","url":"https://handyai.news/modeldrop/gemini-31-pro","published":"2026-02-19"},"launch_score":84,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":9},{"model":"GPT-5.6 Luna","score":84,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.6 Sol, Terra, & Luna","slug":"gpt-56-sol-terra-and-luna","url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","published":"2026-06-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":10},{"model":"GPT-5.4","score":82.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":82.7,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":11},{"model":"GPT-5.2 Pro","score":77.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.6","slug":"claude-opus-46","url":"https://handyai.news/modeldrop/claude-opus-46","published":"2026-02-05"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":12},{"model":"GPT-5.3-Codex","score":77.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-53-codex","rank":13},{"model":"Claude Sonnet 4.6","score":74.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.1 Pro","slug":"gemini-31-pro","url":"https://handyai.news/modeldrop/gemini-31-pro","published":"2026-02-19"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-46","rank":14},{"model":"Qwen3.5","score":69,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":69,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-02-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen35","rank":15},{"model":"ChatGPT agent","score":68.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"first_cited","drop_url":null,"rank":16},{"model":"Claude Opus 4.5","score":67.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":17},{"model":"GPT-5.2","score":65.8,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":65.8,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":18},{"model":"GLM-5","score":62,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/modeldrop/glm-5","published":"2026-02-11"},"launch_score":62,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-02-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-5","rank":19},{"model":"Kimi K2.6","score":61.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-04-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k26","rank":20},{"model":"Kimi K2.5","score":60.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":60.6,"reports":3,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":21},{"model":"Kimi K2 Thinking","score":60.2,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":60.2,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-11-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2-thinking","rank":22},{"model":"GLM-5.1","score":59.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":68,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-03-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-51","rank":23},{"model":"Gemini 3 Pro","score":59.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":24},{"model":"Gemini 3 Pro Deep Research","score":59.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.6","slug":"claude-opus-46","url":"https://handyai.news/modeldrop/claude-opus-46","published":"2026-02-05"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"first_cited","drop_url":null,"rank":25},{"model":"GPT-5","score":54.9,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":54.9,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":26},{"model":"deep research","score":51.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3 & o4-mini","slug":"o3-o4-mini","url":"https://handyai.news/modeldrop/o3-o4-mini","published":"2025-04-16"},"launch_score":null,"reports":1,"maker":null,"open_weights":false,"released":"2025-04-16","dated_by":"first_cited","drop_url":null,"rank":27},{"model":"DeepSeek V3.2","score":51.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":67.6,"reports":4,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-12-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v32","rank":28},{"model":"GPT-5.1 Thinking","score":50.8,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.2","slug":"gpt-52","url":"https://handyai.news/modeldrop/gpt-52","published":"2025-12-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"first_cited","drop_url":null,"rank":29},{"model":"o3","score":49.7,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":49.7,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":30},{"model":"Qwen3.6 35B A3B","score":48.74,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"first_cited","drop_url":null,"rank":31},{"model":"GLM-4.6","score":45.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":32},{"model":"Grok 4 Fast","score":44.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":44.9,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-09-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4-fast","rank":33},{"model":"Nemotron 3 Ultra","score":44.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":44.4,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":34},{"model":"MiniMax-M2","score":44,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":44,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m2","rank":35},{"model":"Grok 4","score":43,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":36},{"model":"Qwen3.5 397B","score":40.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"first_cited","drop_url":null,"rank":37},{"model":"DeepSeek V3.2-Exp","score":40.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":38},{"model":"Nemotron 3.5 Lightning","score":36.97,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":36.97,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-lightning","rank":39},{"model":"DeepSeek V3.1","score":30,"own_drop":true,"independent":false,"setting":"thinking","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":30,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-08-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v31","rank":40},{"model":"o4-mini","score":28.3,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"o3 & o4-mini","slug":"o3-o4-mini","url":"https://handyai.news/modeldrop/o3-o4-mini","published":"2025-04-16"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":41},{"model":"GLM-4.5","score":26.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":26.4,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-07-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-45","rank":42},{"model":"Gemma 4 26B A4B","score":26.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":43},{"model":"Claude Sonnet 4.5","score":24.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.2","slug":"deepseek-v32","url":"https://handyai.news/modeldrop/deepseek-v32","published":"2025-12-01"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":44},{"model":"Claude Opus 4","score":18.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":45},{"model":"Nemotron 3 Nano","score":13.74,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":46},{"model":"DeepSeek R1 0528","score":8.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1-0528","rank":47},{"model":"Kimi K2","score":7.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-07-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2","rank":48},{"model":"GPT-4o","score":1.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3 & o4-mini","slug":"o3-o4-mini","url":"https://handyai.news/modeldrop/o3-o4-mini","published":"2025-04-16"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":49}]},{"key":"terminal-bench-2-1","name":"Terminal-Bench 2.1","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#terminal-bench-2-1","api_url":"https://handyai.news/api/benchmarks/terminal-bench-2-1","image_url":"https://handyai.news/api/benchmarks/terminal-bench-2-1.png","leader":{"model":"MiMo-V2.6-Pro","score":89.9,"maker":"Xiaomi"},"model_count":47,"drop_count":20,"first_date":"2025-08-05","last_date":"2026-11-02","frontier_gain":74.7,"frontier":[{"date":"2025-08-05","model":"gpt-oss-20b","score":15.17,"maker":"openai"},{"date":"2026-02-05","model":"Claude Opus 4.6","score":78.2,"maker":"anthropic"},{"date":"2026-04-23","model":"GPT-5.5","score":85.6,"maker":"openai"},{"date":"2026-06-09","model":"Mythos 5","score":88,"maker":"anthropic"},{"date":"2026-06-26","model":"GPT-5.6 Sol","score":88.8,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":89.1,"maker":"anthropic"},{"date":"2026-09-02","model":"Gemini 3.8 Flash","score":89.4,"maker":"google"},{"date":"2026-09-21","model":"MiMo-V2.6-Pro","score":89.9,"maker":"xiaomi"}],"models":[{"model":"MiMo-V2.6-Pro","score":89.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":89.9,"reports":1,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":1},{"model":"Gemini 3.8 Flash","score":89.4,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":89.4,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":2},{"model":"Claude Opus 5","score":89.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":3},{"model":"GPT-5.6 Sol","score":88.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":88.8,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":4},{"model":"Muse Spark 1.3","score":88.8,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":88.8,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-13","rank":5},{"model":"LongCat-2.5","score":88.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Meituan LongCat-2.5 launch post","slug":"longcat-2-5","url":"https://x.com/Meituan_LongCat/status/2117264311714537633","published":"2026-11-02"},"launch_score":88.4,"reports":1,"maker":{"key":"meituan","name":"Meituan","logo_url":"https://handyai.news/ui/makers/meituan.svg"},"open_weights":true,"released":"2026-11-02","dated_by":"own_drop","drop_url":null,"rank":6},{"model":"Mythos 5","score":88,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":88,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":7},{"model":"GPT-5.6 Terra","score":87.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":8},{"model":"Qwen3.8-Max","score":86.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":86.6,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":9},{"model":"Gemini 3.7 Flash","score":85.8,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":85.8,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-08-13","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-37-flash","rank":10},{"model":"GPT-5.5","score":85.6,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.6 Sol, Terra, & Luna","slug":"gpt-56-sol-terra-and-luna","url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","published":"2026-06-26"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-04-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-55","rank":11},{"model":"Kimi K3","score":85,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":12},{"model":"GPT-5.6 Luna","score":84.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":13},{"model":"Claude Opus 4.8","score":84.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-48","rank":14},{"model":"Fable 5","score":84.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":15},{"model":"GLM-5.3","score":83.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":16},{"model":"Muse Spark 1.2","score":82.9,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":82.9,"reports":3,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-12","rank":17},{"model":"GLM-5.2","score":82.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"first_cited","drop_url":null,"rank":18},{"model":"Qwen3.8-2.4T-A95B","score":82,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":19},{"model":"Grok 4.5","score":81.6,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Muse Spark 1.2","slug":"muse-spark-12","url":"https://handyai.news/modeldrop/muse-spark-12","published":"2026-08-05"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":20},{"model":"Nanbeige4.3-33B","score":80.9,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":80.9,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":21},{"model":"Claude Sonnet 5","score":80.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":22},{"model":"Particle-1","score":79.93,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Particle, \"Introducing Particle-1\"","slug":"particle-1","url":"https://particle.ai/blog/introducing-particle-1","published":"2026-10-28"},"launch_score":79.93,"reports":1,"maker":{"key":"particle","name":"Particle","logo_url":"https://handyai.news/ui/makers/particle.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":null,"rank":23},{"model":"Claude Opus 4.6","score":78.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":24},{"model":"Gemini 3.6 Flash","score":78,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":78,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":25},{"model":"Muse Spark Open","score":76.4,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":76.4,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":26},{"model":"Gemini 3.5 Flash","score":76.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-05-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-35-flash","rank":27},{"model":"Muse Spark 1.1","score":76.2,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark 1.2","slug":"muse-spark-12","url":"https://handyai.news/modeldrop/muse-spark-12","published":"2026-08-05"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-11","rank":28},{"model":"Qwen3.8","score":73,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":73,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":29},{"model":"Kimi K2.6","score":71.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-04-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k26","rank":30},{"model":"Gemini 3.1 Pro","score":70.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":31},{"model":"Qwen3.7-Plus","score":64,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":32},{"model":"Inkling","score":63.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":63.8,"reports":1,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":33},{"model":"Qwen3.6-27B","score":63.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":34},{"model":"GLM-5.1","score":59.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-03-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-51","rank":35},{"model":"Nemotron 3 Ultra","score":56.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":56.4,"reports":2,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":36},{"model":"Muse Glimmer-30B","score":51.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":37},{"model":"Clover-35B-A3B-Preview","score":51.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DiffuSpace, \"Clover-35B-A3B-Preview\"","slug":"clover-35b-a3b-preview","url":"https://diffuspace.ai/blog/clover-35b/","published":"2026-10-28"},"launch_score":51.7,"reports":1,"maker":{"key":"diffuspace","name":"DiffuSpace","logo_url":"https://handyai.news/ui/makers/diffuspace.svg"},"open_weights":true,"released":"2026-10-28","dated_by":"own_drop","drop_url":null,"rank":38},{"model":"Qwen3.5 397B","score":49.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"first_cited","drop_url":null,"rank":39},{"model":"Gemma 4","score":47.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":40},{"model":"Qwen3.6 35B A3B","score":44.38,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"first_cited","drop_url":null,"rank":41},{"model":"SenseNova-Loop-9B-Preview","score":40.45,"own_drop":true,"independent":false,"setting":"","source":{"drop":"SenseNova-Loop-9B-Preview model card","slug":"sensenova-loop-9b-preview","url":"https://huggingface.co/sensenova/SenseNova-Loop-9B-Preview","published":"2026-10-30"},"launch_score":40.45,"reports":1,"maker":{"key":"sensetime","name":"SenseTime","logo_url":"https://handyai.news/ui/makers/sensetime.svg"},"open_weights":true,"released":"2026-10-30","dated_by":"own_drop","drop_url":null,"rank":42},{"model":"Gemma 4 26B A4B","score":37.22,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":43},{"model":"Nemotron 3.5 Lightning","score":24.58,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":24.58,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-lightning","rank":44},{"model":"gpt-oss-20b","score":15.17,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":45},{"model":"LFM3-1.7B","score":12.92,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Liquid AI, \"Introducing LFM3\"","slug":"lfm3-17b","url":"https://www.liquid.ai/blog/lfm3","published":"2026-10-27"},"launch_score":12.92,"reports":1,"maker":{"key":"liquid","name":"Liquid AI","logo_url":"https://handyai.news/ui/makers/liquid.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":46},{"model":"Nemotron 3 Nano","score":8.29,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":47}]},{"key":"arc-agi-2","name":"ARC-AGI-2","about":null,"type":{"key":"reasoning","label":"Reasoning"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#arc-agi-2","api_url":"https://handyai.news/api/benchmarks/arc-agi-2","image_url":"https://handyai.news/api/benchmarks/arc-agi-2.png","leader":{"model":"GPT-5.6 Sol","score":92.5,"maker":"OpenAI"},"model_count":34,"drop_count":20,"first_date":"2025-03-25","last_date":"2026-09-02","frontier_gain":87.6,"frontier":[{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":4.9,"maker":"google"},{"date":"2025-05-22","model":"Claude Opus 4","score":8.6,"maker":"anthropic"},{"date":"2025-07-09","model":"Grok 4","score":16,"maker":"xai"},{"date":"2025-11-12","model":"GPT-5.1","score":17.6,"maker":"openai"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":31.1,"maker":"google"},{"date":"2025-11-24","model":"Claude Opus 4.5","score":37.6,"maker":"anthropic"},{"date":"2025-12-11","model":"GPT-5.2 Pro","score":54.2,"maker":"openai"},{"date":"2026-02-05","model":"Claude Opus 4.6","score":63.3,"maker":"anthropic"},{"date":"2026-02-19","model":"Gemini 3.1 Pro","score":76.5,"maker":"google"},{"date":"2026-03-05","model":"GPT-5.4 Pro","score":83.3,"maker":"openai"},{"date":"2026-06-09","model":"Fable 5","score":89.2,"maker":"anthropic"},{"date":"2026-06-26","model":"GPT-5.6 Sol","score":92.5,"maker":"openai"}],"models":[{"model":"GPT-5.6 Sol","score":92.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":1},{"model":"Claude Opus 5","score":90.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":2},{"model":"Fable 5","score":89.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":3},{"model":"Gemini 3.8 Flash","score":89.2,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":89.2,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":4},{"model":"Gemini 3.7 Flash","score":84.6,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":84.6,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-08-13","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-37-flash","rank":5},{"model":"GPT-5.6 Terra","score":83.9,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":6},{"model":"GPT-5.4 Pro","score":83.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":7},{"model":"Gemini 3.1 Pro","score":76.5,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":77.1,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":8},{"model":"GPT-5.4","score":76.1,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":74,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":9},{"model":"Grok 4.6","score":65.1,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":65.1,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":10},{"model":"Claude Opus 4.6","score":63.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":68.8,"reports":5,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":11},{"model":"Gemini 3.6 Flash","score":60.4,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":60.4,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":12},{"model":"GPT-5.6 Luna","score":59.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":13},{"model":"Claude Sonnet 4.6","score":58.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.1 Pro","slug":"gemini-31-pro","url":"https://handyai.news/modeldrop/gemini-31-pro","published":"2026-02-19"},"launch_score":58.3,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-46","rank":14},{"model":"GPT-5.2 Pro","score":54.2,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":15},{"model":"GPT-5.2","score":52.9,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":52.9,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":16},{"model":"Grok 4.5","score":52.6,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":17},{"model":"Gemini 3 Deep Think","score":45.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.6","slug":"claude-opus-46","url":"https://handyai.news/modeldrop/claude-opus-46","published":"2026-02-05"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"first_cited","drop_url":null,"rank":18},{"model":"Muse Spark","score":42.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":42.5,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-04-08","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark","rank":19},{"model":"Claude Opus 4.5","score":37.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.6","slug":"claude-opus-46","url":"https://handyai.news/modeldrop/claude-opus-46","published":"2026-02-05"},"launch_score":37.6,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":20},{"model":"Gemini 3 Flash","score":33.6,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":33.6,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-flash","rank":21},{"model":"Gemini 3 Pro","score":31.1,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":31.1,"reports":4,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":22},{"model":"GPT-5.1","score":17.6,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Claude Opus 4.5","slug":"claude-opus-45","url":"https://handyai.news/modeldrop/claude-opus-45","published":"2025-11-24"},"launch_score":17.6,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":23},{"model":"GPT-5.1 Thinking","score":17.6,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.2","slug":"gpt-52","url":"https://handyai.news/modeldrop/gpt-52","published":"2025-12-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"first_cited","drop_url":null,"rank":24},{"model":"Grok 4","score":16,"own_drop":false,"independent":true,"setting":"","source":{"drop":"GPT-5.1","slug":"gpt-51","url":"https://handyai.news/modeldrop/gpt-51","published":"2025-11-12"},"launch_score":15.9,"reports":4,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":25},{"model":"Claude Sonnet 4.5","score":13.6,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":13.6,"reports":6,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":26},{"model":"GPT-5","score":9.9,"own_drop":true,"independent":true,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":9.9,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":27},{"model":"Claude Opus 4","score":8.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4","slug":"grok-4","url":"https://handyai.news/modeldrop/grok-4","published":"2025-07-09"},"launch_score":8.6,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":28},{"model":"Claude Sonnet 4","score":5.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":29},{"model":"Grok 4 Fast","score":5.28,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":5.28,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-09-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4-fast","rank":30},{"model":"Gemini 2.5 Pro","score":4.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":null,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":31},{"model":"GPT-5 mini","score":4.44,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-28","dated_by":"first_cited","drop_url":null,"rank":32},{"model":"o3","score":2.9,"own_drop":true,"independent":true,"setting":"","source":{"drop":"o3 & o4-mini","slug":"o3-o4-mini","url":"https://handyai.news/modeldrop/o3-o4-mini","published":"2025-04-16"},"launch_score":2.9,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":33},{"model":"o4-mini","score":2.3,"own_drop":false,"independent":true,"setting":"medium","source":{"drop":"o3 & o4-mini","slug":"o3-o4-mini","url":"https://handyai.news/modeldrop/o3-o4-mini","published":"2025-04-16"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":34}]},{"key":"gdpval-aa","name":"GDPval-AA","about":"Professional work products, graded head to head (Elo).","type":{"key":"work","label":"Work"},"unit":"Elo","lower_is_better":false,"url":"https://handyai.news/benchmarks#gdpval-aa","api_url":"https://handyai.news/api/benchmarks/gdpval-aa","image_url":"https://handyai.news/api/benchmarks/gdpval-aa.png","leader":{"model":"Claude Opus 5.5","score":1866,"maker":"Anthropic"},"model_count":39,"drop_count":19,"first_date":"2025-10-15","last_date":"2026-10-23","frontier_gain":1131,"frontier":[{"date":"2025-10-15","model":"Claude Haiku 4.5","score":735,"maker":"anthropic"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":1201,"maker":"google"},{"date":"2025-11-24","model":"Claude Opus 4.5","score":1416,"maker":"anthropic"},{"date":"2025-12-11","model":"GPT-5.2","score":1462,"maker":"openai"},{"date":"2026-02-05","model":"Claude Opus 4.6","score":1606,"maker":"anthropic"},{"date":"2026-02-17","model":"Claude Sonnet 4.6","score":1633,"maker":"anthropic"},{"date":"2026-06-09","model":"Fable 5","score":1743,"maker":"anthropic"},{"date":"2026-08-14","model":"GLM-5.3","score":1769,"maker":"zai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":1866,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":1866,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":1846,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"Muse Wonder Full","score":1857,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":2},{"model":"Kimi K3.1 Pro","score":1846,"own_drop":true,"independent":true,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":1846,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":3},{"model":"Claude Sonnet 5.5","score":1838,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":1844,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":4},{"model":"Muse Wonder","score":1770,"own_drop":true,"independent":true,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":1770,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":5},{"model":"GLM-5.3","score":1769,"own_drop":true,"independent":true,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":1769,"reports":3,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":6},{"model":"Fable 5.1","score":1754,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":1853,"reports":6,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":7},{"model":"Muse Spark 1.3","score":1754,"own_drop":true,"independent":true,"setting":"max","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":1754,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-13","rank":8},{"model":"DeepSeek V4.1 Flash 1017","score":1746,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":9},{"model":"Fable 5","score":1743,"own_drop":false,"independent":true,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":10},{"model":"Claude Opus 5","score":1708,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Opus 5.5","slug":"claude-opus-55","url":"https://handyai.news/modeldrop/claude-opus-55","published":"2026-09-22"},"launch_score":null,"reports":7,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":11},{"model":"Grok 4.7","score":1695,"own_drop":true,"independent":true,"setting":"xhigh","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":1695,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":12},{"model":"MiMo-V2.6-Pro","score":1673,"own_drop":true,"independent":true,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":1673,"reports":1,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":13},{"model":"Claude Sonnet 4.6","score":1633,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Gemini 3.1 Pro","slug":"gemini-31-pro","url":"https://handyai.news/modeldrop/gemini-31-pro","published":"2026-02-19"},"launch_score":1633,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-46","rank":14},{"model":"Claude Haiku 5.5","score":1620,"own_drop":true,"independent":true,"setting":"max","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":1620,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":15},{"model":"Muse Spark 1.2","score":1615,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":1631,"reports":3,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-12","rank":16},{"model":"Qwen3.8-2.4T-A95B","score":1613,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":17},{"model":"Claude Opus 4.6","score":1606,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Gemini 3.1 Pro","slug":"gemini-31-pro","url":"https://handyai.news/modeldrop/gemini-31-pro","published":"2026-02-19"},"launch_score":1606,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":18},{"model":"Grok 4.6","score":1605,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":1753,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":19},{"model":"GPT-5.6 Sol","score":1588,"own_drop":false,"independent":true,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":null,"reports":6,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":20},{"model":"GPT-5.6 Luna","score":1584,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":21},{"model":"GPT-6 Astra","score":1573,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":22},{"model":"Gemini 3.8 Flash","score":1545,"own_drop":true,"independent":true,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":1545,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":23},{"model":"Kimi K3","score":1537,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":3,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":24},{"model":"GPT-5.6 Terra","score":1528,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":25},{"model":"Grok 4.5","score":1526,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":3,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":26},{"model":"Gemini 3.7 Flash","score":1525,"own_drop":true,"independent":true,"setting":"high","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":1525,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-08-13","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-37-flash","rank":27},{"model":"GPT-6 Sol","score":1487,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Sonnet 5.5","slug":"claude-sonnet-55","url":"https://handyai.news/modeldrop/claude-sonnet-55","published":"2026-09-28"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":28},{"model":"GPT-5.2","score":1462,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Gemini 3.1 Pro","slug":"gemini-31-pro","url":"https://handyai.news/modeldrop/gemini-31-pro","published":"2026-02-19"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":29},{"model":"Claude Sonnet 5","score":1449,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Sonnet 5.5","slug":"claude-sonnet-55","url":"https://handyai.news/modeldrop/claude-sonnet-55","published":"2026-09-28"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":30},{"model":"GPT-6 Luna","score":1437,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":31},{"model":"Mistral Large 4","score":1424,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":32},{"model":"Gemini 3.6 Flash","score":1422,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":1421,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":33},{"model":"Claude Opus 4.5","score":1416,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Opus 4.6","slug":"claude-opus-46","url":"https://handyai.news/modeldrop/claude-opus-46","published":"2026-02-05"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":34},{"model":"Muse Spark Open","score":1401,"own_drop":true,"independent":true,"setting":"xhigh","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":1401,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":35},{"model":"Gemini 3.5 Flash","score":1349,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-05-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-35-flash","rank":36},{"model":"Gemini 3.1 Pro","score":1317,"own_drop":true,"independent":true,"setting":"max","source":{"drop":"Gemini 3.1 Pro","slug":"gemini-31-pro","url":"https://handyai.news/modeldrop/gemini-31-pro","published":"2026-02-19"},"launch_score":1317,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":37},{"model":"Gemini 3 Pro","score":1201,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":38},{"model":"Claude Haiku 4.5","score":735,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":39}]},{"key":"terminal-bench-4-0","name":"Terminal-Bench 4.0","about":"Multi-hour work in a real terminal.","type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#terminal-bench-4-0","api_url":"https://handyai.news/api/benchmarks/terminal-bench-4-0","image_url":"https://handyai.news/api/benchmarks/terminal-bench-4-0.png","leader":{"model":"Claude Sonnet 5.5","score":70.6,"maker":"Anthropic"},"model_count":37,"drop_count":19,"first_date":"2025-10-15","last_date":"2026-11-04","frontier_gain":70.6,"frontier":[{"date":"2025-10-15","model":"Claude Haiku 4.5","score":0,"maker":"anthropic"},{"date":"2026-06-26","model":"GPT-5.6 Sol","score":39.9,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":51.8,"maker":"anthropic"},{"date":"2026-09-01","model":"Fable 5.1","score":57.9,"maker":"anthropic"},{"date":"2026-09-03","model":"Mythos 5.1","score":60.9,"maker":"anthropic"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":66.4,"maker":"anthropic"},{"date":"2026-09-28","model":"Claude Sonnet 5.5","score":70.6,"maker":"anthropic"}],"models":[{"model":"Claude Sonnet 5.5","score":70.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":70.6,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":1},{"model":"Claude Opus 5.5","score":66.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":66.4,"reports":5,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2},{"model":"Kimi K3.1 Pro","score":65.7,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":65.7,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":3},{"model":"Mythos 5.1","score":60.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-6 Astra","slug":"gpt-6-astra","url":"https://handyai.news/modeldrop/gpt-6-astra","published":"2026-09-03"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"first_cited","drop_url":null,"rank":4},{"model":"GPT-6 Astra","score":58.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":57.7,"reports":7,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":5},{"model":"Fable 5.1","score":57.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":60.9,"reports":7,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":6},{"model":"Gemini 4 Argon","score":57.4,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Argon","slug":"gemini-4-argon","url":"https://handyai.news/modeldrop/gemini-4-argon","published":"2026-10-01"},"launch_score":57.4,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":7},{"model":"Muse Wonder Full","score":57.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":8},{"model":"GPT-6.1 Sol","score":55.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":9},{"model":"Muse Wonder","score":52,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":52,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":10},{"model":"Claude Opus 5","score":51.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":5,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":11},{"model":"GPT-6.1 Terra","score":50.3,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":50.3,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":12},{"model":"Hy4","score":44.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Tencent Hunyuan Hy4 launch post","slug":"hy4","url":"https://x.com/TencentHunyuan/status/2117870250838630539","published":"2026-11-04"},"launch_score":44.4,"reports":1,"maker":{"key":"tencent","name":"Tencent","logo_url":"https://handyai.news/ui/makers/tencent.svg"},"open_weights":true,"released":"2026-11-04","dated_by":"own_drop","drop_url":null,"rank":13},{"model":"DeepSeek V4.1 Flash 1017","score":42.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":42.3,"reports":3,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":14},{"model":"GLM-5.3","score":41.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":4,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":15},{"model":"GPT-5.6 Sol","score":39.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":6,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":16},{"model":"Claude Haiku 5.5","score":39.2,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":39.2,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":17},{"model":"LongCat-2.5","score":38.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Meituan LongCat-2.5 launch post","slug":"longcat-2-5","url":"https://x.com/Meituan_LongCat/status/2117264311714537633","published":"2026-11-02"},"launch_score":38.5,"reports":1,"maker":{"key":"meituan","name":"Meituan","logo_url":"https://handyai.news/ui/makers/meituan.svg"},"open_weights":true,"released":"2026-11-02","dated_by":"own_drop","drop_url":null,"rank":18},{"model":"MiniMax M3.1 Flash","score":38.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":38.1,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":19},{"model":"Grok 4.7","score":37.6,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":37.6,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":20},{"model":"MiMo-V2.6-Pro","score":34.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":34.9,"reports":1,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":21},{"model":"DeepSeek V4.1 Flash","score":31.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":22},{"model":"Mistral Large 4","score":28.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":28.3,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":23},{"model":"Nanbeige4.3-33B","score":23.8,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":23.8,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":24},{"model":"GPT-5.6 Terra","score":21.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":25},{"model":"Grok 4.6","score":20.3,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":26},{"model":"Gemini 3.8 Flash","score":19.1,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":19.1,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":27},{"model":"GPT-6 Luna","score":16.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":28},{"model":"Kimi K3","score":12.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":4,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":29},{"model":"Claude Sonnet 5","score":10.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 5.5","slug":"claude-sonnet-55","url":"https://handyai.news/modeldrop/claude-sonnet-55","published":"2026-09-28"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":30},{"model":"DeepSeek V4 Pro 0813","score":10,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"first_cited","drop_url":null,"rank":31},{"model":"Qwen3.8","score":6.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":32},{"model":"Fern 0.1","score":6.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mirendil, \"Introducing Fern\"","slug":"fern-01","url":"https://mirendil.com/news/introducing-fern/","published":"2026-10-26"},"launch_score":6.1,"reports":1,"maker":{"key":"mirendil","name":"Mirendil","logo_url":"https://handyai.news/ui/makers/mirendil.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":null,"rank":33},{"model":"Muse Glimmer-30B","score":1.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":34},{"model":"Claude Haiku 4.5","score":0,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":35},{"model":"Gemma 4","score":0,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":36},{"model":"Nemotron 3 Ultra","score":0,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mirendil, \"Introducing Fern\"","slug":"fern-01","url":"https://mirendil.com/news/introducing-fern/","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":37}]},{"key":"mmmu","name":"MMMU","about":null,"type":{"key":"multimodal","label":"Multimodal"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#mmmu","api_url":"https://handyai.news/api/benchmarks/mmmu","image_url":"https://handyai.news/api/benchmarks/mmmu.png","leader":{"model":"Gemini 3 Pro","score":87.2,"maker":"Google DeepMind"},"model_count":32,"drop_count":17,"first_date":"2024-12-05","last_date":"2026-02-16","frontier_gain":9.6,"frontier":[{"date":"2024-12-05","model":"o1","score":77.6,"maker":"openai"},{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":82,"maker":"google"},{"date":"2025-04-16","model":"o3","score":82.9,"maker":"openai"},{"date":"2025-08-07","model":"GPT-5","score":84.2,"maker":"openai"},{"date":"2025-11-12","model":"GPT-5.1","score":85.4,"maker":"openai"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":87.2,"maker":"google"}],"models":[{"model":"Gemini 3 Pro","score":87.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":1},{"model":"GPT-5.2","score":86.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":2},{"model":"GPT-5.1","score":85.4,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Claude Opus 4.5","slug":"claude-opus-45","url":"https://handyai.news/modeldrop/claude-opus-45","published":"2025-11-24"},"launch_score":85.4,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":3},{"model":"Qwen3.5","score":85,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":85,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-02-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen35","rank":4},{"model":"Kimi K2.5","score":84.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":5},{"model":"GPT-5","score":84.2,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.1","slug":"gpt-51","url":"https://handyai.news/modeldrop/gpt-51","published":"2025-11-12"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":6},{"model":"o3","score":82.9,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":82.9,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":7},{"model":"Gemini 2.5 Pro","score":82,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":81.7,"reports":4,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":8},{"model":"o4-mini","score":81.6,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":9},{"model":"Claude Opus 4.5","score":80.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":80.7,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":10},{"model":"Gemini 2.5 Pro Preview","score":79.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"first_cited","drop_url":null,"rank":11},{"model":"Claude Sonnet 4.5","score":77.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.5","slug":"claude-opus-45","url":"https://handyai.news/modeldrop/claude-opus-45","published":"2025-11-24"},"launch_score":77.8,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":12},{"model":"o1","score":77.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3 & o4-mini","slug":"o3-o4-mini","url":"https://handyai.news/modeldrop/o3-o4-mini","published":"2025-04-16"},"launch_score":77.3,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":13},{"model":"Claude Opus 4.1","score":77.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":77.1,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":14},{"model":"Gemini 2.5 Flash","score":76.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":76.7,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-flash","rank":15},{"model":"Claude Opus 4","score":76.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":76.5,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":16},{"model":"Grok 3","score":76,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Pro","slug":"gemini-25-pro","url":"https://handyai.news/modeldrop/gemini-25-pro","published":"2025-03-25"},"launch_score":76,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-3","rank":17},{"model":"Grok 3 Beta","score":76,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"first_cited","drop_url":null,"rank":18},{"model":"Gemini 2.0 Flash Thinking","score":75.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 3","slug":"grok-3","url":"https://handyai.news/modeldrop/grok-3","published":"2025-02-17"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"first_cited","drop_url":null,"rank":19},{"model":"Claude 3.7 Sonnet","score":75,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":75,"reports":5,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":20},{"model":"GPT-4.1","score":74.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":21},{"model":"GPT-4.5","score":74.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Pro","slug":"gemini-25-pro","url":"https://handyai.news/modeldrop/gemini-25-pro","published":"2025-03-25"},"launch_score":74.4,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":22},{"model":"Claude Sonnet 4","score":74.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":23},{"model":"Claude Haiku 4.5","score":73.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":73.2,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":24},{"model":"Llama 4 Maverick","score":71.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":73.4,"reports":3,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":25},{"model":"Gemini 2.0 Flash","score":71.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":70.7,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":26},{"model":"Claude 3.5 Sonnet","score":70.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude 3.7 Sonnet","slug":"claude-37-sonnet","url":"https://handyai.news/modeldrop/claude-37-sonnet","published":"2025-02-24"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":27},{"model":"Llama 4 Scout","score":69.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":69.4,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":28},{"model":"GPT-4o","score":68.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":29},{"model":"Mistral Medium 3","score":66.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":66.1,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2025-05-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-medium-3","rank":30},{"model":"Gemini 1.5 Pro","score":65.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":31},{"model":"Gemini 1.5 Flash","score":62.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"first_cited","drop_url":null,"rank":32}]},{"key":"arc-agi-1","name":"ARC-AGI-1","about":null,"type":{"key":"reasoning","label":"Reasoning"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#arc-agi-1","api_url":"https://handyai.news/api/benchmarks/arc-agi-1","image_url":"https://handyai.news/api/benchmarks/arc-agi-1.png","leader":{"model":"Gemini 3.1 Pro","score":98,"maker":"Google DeepMind"},"model_count":28,"drop_count":17,"first_date":"2024-12-05","last_date":"2026-08-13","frontier_gain":63,"frontier":[{"date":"2024-12-05","model":"o1","score":35,"maker":"openai"},{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":41,"maker":"google"},{"date":"2025-04-16","model":"o3","score":53,"maker":"openai"},{"date":"2025-07-09","model":"Grok 4","score":66.7,"maker":"xai"},{"date":"2025-11-12","model":"GPT-5.1","score":72.83,"maker":"openai"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":75,"maker":"google"},{"date":"2025-11-24","model":"Claude Opus 4.5","score":80,"maker":"anthropic"},{"date":"2025-12-11","model":"GPT-5.2","score":86.2,"maker":"openai"},{"date":"2026-02-05","model":"Claude Opus 4.6","score":94,"maker":"anthropic"},{"date":"2026-02-19","model":"Gemini 3.1 Pro","score":98,"maker":"google"}],"models":[{"model":"Gemini 3.1 Pro","score":98,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemini 3.1 Pro","slug":"gemini-31-pro","url":"https://handyai.news/modeldrop/gemini-31-pro","published":"2026-02-19"},"launch_score":98,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":1},{"model":"GPT-5.6 Terra","score":96.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":2},{"model":"Gemini 3.7 Flash","score":95.5,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":95.5,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-08-13","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-37-flash","rank":3},{"model":"Claude Opus 4.6","score":94,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Gemini 3.1 Pro","slug":"gemini-31-pro","url":"https://handyai.news/modeldrop/gemini-31-pro","published":"2026-02-19"},"launch_score":93,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":4},{"model":"Gemini 3.6 Flash","score":91.2,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":5},{"model":"GPT-5.2","score":86.2,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Gemini 3.1 Pro","slug":"gemini-31-pro","url":"https://handyai.news/modeldrop/gemini-31-pro","published":"2026-02-19"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":6},{"model":"Claude Sonnet 4.6","score":86,"own_drop":true,"independent":true,"setting":"max","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":86,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-46","rank":7},{"model":"Gemini 3 Flash","score":84.7,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":84.7,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-flash","rank":8},{"model":"Claude Opus 4.5","score":80,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Claude Opus 4.5","slug":"claude-opus-45","url":"https://handyai.news/modeldrop/claude-opus-45","published":"2025-11-24"},"launch_score":80,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":9},{"model":"Gemini 3 Pro","score":75,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":75,"reports":5,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":10},{"model":"GPT-5.1","score":72.83,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Claude Opus 4.5","slug":"claude-opus-45","url":"https://handyai.news/modeldrop/claude-opus-45","published":"2025-11-24"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":11},{"model":"Grok 4","score":66.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":66.7,"reports":3,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":12},{"model":"GPT-5","score":65.7,"own_drop":true,"independent":true,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":65.7,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":13},{"model":"Claude Sonnet 4.5","score":63.7,"own_drop":false,"independent":true,"setting":"thinking","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":14},{"model":"GPT-5 mini","score":54.33,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-28","dated_by":"first_cited","drop_url":null,"rank":15},{"model":"o3","score":53,"own_drop":true,"independent":true,"setting":"","source":{"drop":"o3 & o4-mini","slug":"o3-o4-mini","url":"https://handyai.news/modeldrop/o3-o4-mini","published":"2025-04-16"},"launch_score":53,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":16},{"model":"Grok 4 Fast","score":48.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":48.5,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-09-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4-fast","rank":17},{"model":"Claude Haiku 4.5","score":47.67,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":47.67,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":18},{"model":"Gemini 2.5 Pro","score":41,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4","slug":"grok-4","url":"https://handyai.news/modeldrop/grok-4","published":"2025-07-09"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":19},{"model":"Claude Opus 4","score":35.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4","slug":"grok-4","url":"https://handyai.news/modeldrop/grok-4","published":"2025-07-09"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":20},{"model":"o1","score":35,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"DeepSeek R1","slug":"deepseek-r1","url":"https://handyai.news/modeldrop/deepseek-r1","published":"2025-01-20"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":21},{"model":"o3-mini","score":34.5,"own_drop":true,"independent":true,"setting":"","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":34.5,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":22},{"model":"Claude 3.7 Sonnet","score":28.6,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Claude 3.7 Sonnet","slug":"claude-37-sonnet","url":"https://handyai.news/modeldrop/claude-37-sonnet","published":"2025-02-24"},"launch_score":28.6,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":23},{"model":"DeepSeek R1","score":15.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude 3.7 Sonnet","slug":"claude-37-sonnet","url":"https://handyai.news/modeldrop/claude-37-sonnet","published":"2025-02-24"},"launch_score":15.8,"reports":3,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":24},{"model":"o1-mini","score":14,"own_drop":false,"independent":true,"setting":"","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-20","dated_by":"first_cited","drop_url":null,"rank":25},{"model":"GPT-4.5","score":10.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":10.3,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":26},{"model":"GPT-4.1","score":5.5,"own_drop":true,"independent":true,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":5.5,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":27},{"model":"GPT-4o","score":4.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":28}]},{"key":"aider-polyglot","name":"Aider Polyglot","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#aider-polyglot","api_url":"https://handyai.news/api/benchmarks/aider-polyglot","image_url":"https://handyai.news/api/benchmarks/aider-polyglot.png","leader":{"model":"GPT-5","score":88,"maker":"OpenAI"},"model_count":27,"drop_count":17,"first_date":"2024-12-05","last_date":"2025-08-21","frontier_gain":26.3,"frontier":[{"date":"2024-12-05","model":"o1","score":61.7,"maker":"openai"},{"date":"2025-02-24","model":"Claude 3.7 Sonnet","score":64.9,"maker":"anthropic"},{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":76.9,"maker":"google"},{"date":"2025-04-16","model":"o3","score":79.6,"maker":"openai"},{"date":"2025-06-10","model":"o3-pro","score":84.9,"maker":"openai"},{"date":"2025-08-07","model":"GPT-5","score":88,"maker":"openai"}],"models":[{"model":"GPT-5","score":88,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":88,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":1},{"model":"o3-pro","score":84.9,"own_drop":true,"independent":true,"setting":"","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":84.9,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-06-10","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-pro","rank":2},{"model":"o3","score":79.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":3},{"model":"Gemini 2.5 Pro","score":76.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":68.6,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":4},{"model":"DeepSeek V3.1","score":76.3,"own_drop":true,"independent":false,"setting":"thinking","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":76.3,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-08-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v31","rank":5},{"model":"DeepSeek R1 0528","score":71.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":71.6,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1-0528","rank":6},{"model":"Claude Opus 4","score":70.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":72,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":7},{"model":"o4-mini","score":68.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":8},{"model":"Qwen3 235B","score":65,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"first_cited","drop_url":null,"rank":9},{"model":"Claude 3.7 Sonnet","score":64.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Pro","slug":"gemini-25-pro","url":"https://handyai.news/modeldrop/gemini-25-pro","published":"2025-03-25"},"launch_score":64.9,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":10},{"model":"o1","score":61.7,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":61.7,"reports":6,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":11},{"model":"Claude Sonnet 4","score":61.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":12},{"model":"o3-mini","score":60.4,"own_drop":true,"independent":true,"setting":"","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":60.4,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":13},{"model":"Kimi K2","score":60,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":60,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-07-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2","rank":14},{"model":"Qwen3","score":59.6,"own_drop":true,"independent":true,"setting":"no thinking","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":59.6,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-04-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3","rank":15},{"model":"DeepSeek R1","score":57,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":56.9,"reports":6,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":16},{"model":"DeepSeek V3-0324","score":55.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"first_cited","drop_url":null,"rank":17},{"model":"DeepSeek V3 0324","score":55.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-14","dated_by":"first_cited","drop_url":null,"rank":18},{"model":"GPT-4.1","score":52.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":52.9,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":19},{"model":"Claude 3.5 Sonnet","score":51.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude 3.7 Sonnet","slug":"claude-37-sonnet","url":"https://handyai.news/modeldrop/claude-37-sonnet","published":"2025-02-24"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":20},{"model":"Gemini 2.5 Flash","score":51.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":51.1,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-flash","rank":21},{"model":"DeepSeek V3","score":48.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":48.4,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v3","rank":22},{"model":"GPT-4.5","score":44.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":44.9,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":23},{"model":"o1-mini","score":32.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"DeepSeek R1","slug":"deepseek-r1","url":"https://handyai.news/modeldrop/deepseek-r1","published":"2025-01-20"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-20","dated_by":"first_cited","drop_url":null,"rank":24},{"model":"Gemini 2.0 Flash","score":22.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":25},{"model":"GPT-4o","score":18.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":null,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":26},{"model":"Llama 4 Maverick","score":15.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":15.6,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":27}]},{"key":"automationbench","name":"AutomationBench","about":"Real business automations (Zapier).","type":{"key":"work","label":"Work"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#automationbench","api_url":"https://handyai.news/api/benchmarks/automationbench","image_url":"https://handyai.news/api/benchmarks/automationbench.png","leader":{"model":"MiniMax M3.1 Flash","score":62.2,"maker":"MiniMax"},"model_count":31,"drop_count":16,"first_date":"2026-04-02","last_date":"2026-11-04","frontier_gain":52.5,"frontier":[{"date":"2026-04-02","model":"Gemma 4","score":9.7,"maker":"google"},{"date":"2026-06-09","model":"Fable 5","score":46.2,"maker":"anthropic"},{"date":"2026-07-16","model":"Kimi K3","score":46.7,"maker":"moonshot"},{"date":"2026-07-24","model":"Claude Opus 5","score":50.3,"maker":"anthropic"},{"date":"2026-08-14","model":"DeepSeek V4 Pro 0813","score":56.7,"maker":"deepseek"},{"date":"2026-10-06","model":"Mistral Large 4","score":59.9,"maker":"mistral"},{"date":"2026-10-17","model":"DeepSeek V4.1 Flash 1017","score":60.5,"maker":"deepseek"},{"date":"2026-10-19","model":"MiniMax M3.1 Flash","score":62.2,"maker":"minimax"}],"models":[{"model":"MiniMax M3.1 Flash","score":62.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":62.2,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":1},{"model":"DeepSeek V4.1 Flash 1017","score":60.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":60.5,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":2},{"model":"Mistral Large 4","score":59.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":59.9,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":3},{"model":"DeepSeek V4 Pro 0813","score":56.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"first_cited","drop_url":null,"rank":4},{"model":"Hy4","score":56.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Tencent Hunyuan Hy4 launch post","slug":"hy4","url":"https://x.com/TencentHunyuan/status/2117870250838630539","published":"2026-11-04"},"launch_score":56.2,"reports":1,"maker":{"key":"tencent","name":"Tencent","logo_url":"https://handyai.news/ui/makers/tencent.svg"},"open_weights":true,"released":"2026-11-04","dated_by":"own_drop","drop_url":null,"rank":5},{"model":"DeepSeek V4.1 Flash","score":54.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":6},{"model":"Nanbeige4.3-33B","score":54.2,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":54.2,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":7},{"model":"MiMo-V2.6-Pro","score":53.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":53.1,"reports":1,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":8},{"model":"LongCat-2.5","score":51.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Meituan LongCat-2.5 launch post","slug":"longcat-2-5","url":"https://x.com/Meituan_LongCat/status/2117264311714537633","published":"2026-11-02"},"launch_score":51.8,"reports":1,"maker":{"key":"meituan","name":"Meituan","logo_url":"https://handyai.news/ui/makers/meituan.svg"},"open_weights":true,"released":"2026-11-02","dated_by":"own_drop","drop_url":null,"rank":9},{"model":"Gemini 4 Argon","score":51.3,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Argon","slug":"gemini-4-argon","url":"https://handyai.news/modeldrop/gemini-4-argon","published":"2026-10-01"},"launch_score":51.3,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":10},{"model":"Claude Opus 5","score":50.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":5,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":11},{"model":"Muse Spark 1.3","score":49.6,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":49.6,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-13","rank":12},{"model":"GLM-5.3","score":48.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":48.2,"reports":4,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":13},{"model":"Kimi K3","score":46.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":5,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":14},{"model":"Fable 5","score":46.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":15},{"model":"GPT-5.6 Sol","score":45.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":16},{"model":"Claude Sonnet 5.5","score":44.7,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Claude Sonnet 5.5","slug":"claude-sonnet-55","url":"https://handyai.news/modeldrop/claude-sonnet-55","published":"2026-09-28"},"launch_score":44.7,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":17},{"model":"Claude Opus 5.5","score":42.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 4 Argon","slug":"gemini-4-argon","url":"https://handyai.news/modeldrop/gemini-4-argon","published":"2026-10-01"},"launch_score":40,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":18},{"model":"GPT-6 Astra","score":41.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Argon","slug":"gemini-4-argon","url":"https://handyai.news/modeldrop/gemini-4-argon","published":"2026-10-01"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":19},{"model":"Qwen3.8","score":38.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":20},{"model":"GPT-6.1 Sol","score":36.1,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Sol","slug":"gpt-61-sol","url":"https://handyai.news/modeldrop/gpt-61-sol","published":"2026-09-29"},"launch_score":36.1,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":21},{"model":"GPT-6 Sol","score":33.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-6.1 Sol","slug":"gpt-61-sol","url":"https://handyai.news/modeldrop/gpt-61-sol","published":"2026-09-29"},"launch_score":33.2,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":22},{"model":"Fable 5.1","score":31.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Argon","slug":"gemini-4-argon","url":"https://handyai.news/modeldrop/gemini-4-argon","published":"2026-10-01"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":23},{"model":"Gemini 3.7 Flash","score":30.4,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":30.4,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-08-13","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-37-flash","rank":24},{"model":"GPT-5.6 Terra","score":23.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":25},{"model":"GPT-6 Luna","score":20.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6 Sol & GPT-6 Luna","slug":"gpt-6-sol-and-gpt-6-luna","url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","published":"2026-09-22"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":26},{"model":"Gemini 3.6 Flash","score":17,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":27},{"model":"Muse Glimmer-30B","score":15.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":28},{"model":"SenseNova-Loop-9B-Preview","score":13.72,"own_drop":true,"independent":false,"setting":"","source":{"drop":"SenseNova-Loop-9B-Preview model card","slug":"sensenova-loop-9b-preview","url":"https://huggingface.co/sensenova/SenseNova-Loop-9B-Preview","published":"2026-10-30"},"launch_score":13.72,"reports":1,"maker":{"key":"sensetime","name":"SenseTime","logo_url":"https://handyai.news/ui/makers/sensetime.svg"},"open_weights":true,"released":"2026-10-30","dated_by":"own_drop","drop_url":null,"rank":29},{"model":"Claude Sonnet 5","score":10.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 5.5","slug":"claude-sonnet-55","url":"https://handyai.news/modeldrop/claude-sonnet-55","published":"2026-09-28"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":30},{"model":"Gemma 4","score":9.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":31}]},{"key":"swe-bench-pro","name":"SWE-bench Pro","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#swe-bench-pro","api_url":"https://handyai.news/api/benchmarks/swe-bench-pro","image_url":"https://handyai.news/api/benchmarks/swe-bench-pro.png","leader":{"model":"Fable 5","score":80,"maker":"Anthropic"},"model_count":36,"drop_count":15,"first_date":"2025-10-15","last_date":"2026-10-24","frontier_gain":44.8,"frontier":[{"date":"2025-10-15","model":"Claude Haiku 4.5","score":35.2,"maker":"anthropic"},{"date":"2025-11-12","model":"GPT-5.1","score":50.8,"maker":"openai"},{"date":"2025-12-11","model":"GPT-5.2","score":55.6,"maker":"openai"},{"date":"2025-12-18","model":"GPT-5.2-Codex","score":56.4,"maker":"openai"},{"date":"2026-02-05","model":"GPT-5.3-Codex","score":56.8,"maker":"openai"},{"date":"2026-03-05","model":"GPT-5.4","score":57.7,"maker":"openai"},{"date":"2026-03-27","model":"GLM-5.1","score":58.4,"maker":"zai"},{"date":"2026-04-07","model":"Claude Mythos Preview","score":77.8,"maker":"anthropic"},{"date":"2026-06-09","model":"Fable 5","score":80,"maker":"anthropic"}],"models":[{"model":"Fable 5","score":80,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":1},{"model":"Claude Mythos Preview","score":77.8,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Claude Mythos Preview","slug":"claude-mythos-preview","url":"https://handyai.news/modeldrop/claude-mythos-preview","published":"2026-04-07"},"launch_score":77.8,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":2},{"model":"Claude Opus 4.8","score":69.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-48","rank":3},{"model":"Qwen3.8-Max","score":67.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":67.7,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":4},{"model":"Grok 4.5","score":64.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":5},{"model":"GPT-5.6 Sol","score":64.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":6},{"model":"Claude Sonnet 5","score":63.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":7},{"model":"GPT-5.6 Luna","score":62.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":8},{"model":"GLM-5.2","score":62.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"first_cited","drop_url":null,"rank":9},{"model":"Qwen3.8","score":61.7,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":61.7,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":10},{"model":"Qwen3.8-Flash-Next","score":60.37,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"first_cited","drop_url":null,"rank":11},{"model":"Nanbeige4.3-33B","score":60.3,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":60.3,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":12},{"model":"GLM-5.3-Flash","score":59.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"first_cited","drop_url":null,"rank":13},{"model":"Gemini 3.6 Flash","score":58.7,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":58.7,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":14},{"model":"Kimi K2.6","score":58.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-04-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k26","rank":15},{"model":"GLM-5.1","score":58.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":58.4,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-03-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-51","rank":16},{"model":"GPT-5.4","score":57.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":57.7,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":17},{"model":"Qwen3.7-Plus","score":57.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":18},{"model":"GPT-5.3-Codex","score":56.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":56.8,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-53-codex","rank":19},{"model":"GPT-5.2-Codex","score":56.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.3-Codex","slug":"gpt-53-codex","url":"https://handyai.news/modeldrop/gpt-53-codex","published":"2026-02-05"},"launch_score":56.4,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52-codex","rank":20},{"model":"GPT-5.2","score":55.6,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":55.6,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":21},{"model":"Gemini 3.5 Flash","score":55.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-05-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-35-flash","rank":22},{"model":"Inkling","score":54.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":54.3,"reports":1,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":23},{"model":"Gemini 3.1 Pro","score":54.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Claude Mythos Preview","slug":"claude-mythos-preview","url":"https://handyai.news/modeldrop/claude-mythos-preview","published":"2026-04-07"},"launch_score":54.2,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":24},{"model":"Kimi K2.5","score":53.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":25},{"model":"Qwen3.6-27B","score":53.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":26},{"model":"Claude Opus 4.6","score":53.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":27},{"model":"MAI-Thinking-1","score":52.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":52.8,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-thinking-1","rank":28},{"model":"Nemotron 3.5 Super VL","score":51.78,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":51.78,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":29},{"model":"MAI-Code-1-Flash","score":51.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":51.2,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-06-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-code-1-flash","rank":30},{"model":"Muse Glimmer-30B","score":51.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":31},{"model":"Gemma 4","score":51.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":32},{"model":"GPT-5.1 Thinking","score":50.8,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.2","slug":"gpt-52","url":"https://handyai.news/modeldrop/gpt-52","published":"2025-12-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"first_cited","drop_url":null,"rank":33},{"model":"GPT-5.1","score":50.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":34},{"model":"Nemotron 3 Ultra","score":49.66,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":2,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":35},{"model":"Claude Haiku 4.5","score":35.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":36}]},{"key":"humanitys-last-exam-with-tools","name":"Humanity's Last Exam (with tools)","about":"Expert-level questions, with search and code.","type":{"key":"reasoning","label":"Reasoning"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#humanitys-last-exam-with-tools","api_url":"https://handyai.news/api/benchmarks/humanitys-last-exam-with-tools","image_url":"https://handyai.news/api/benchmarks/humanitys-last-exam-with-tools.png","leader":{"model":"Kimi K3.1 Pro","score":70,"maker":"Moonshot AI"},"model_count":40,"drop_count":14,"first_date":"2025-04-16","last_date":"2026-10-27","frontier_gain":45.1,"frontier":[{"date":"2025-04-16","model":"o3","score":24.9,"maker":"openai"},{"date":"2025-07-09","model":"Grok 4","score":41,"maker":"xai"},{"date":"2025-08-07","model":"GPT-5","score":41.7,"maker":"openai"},{"date":"2025-11-06","model":"Kimi K2 Thinking","score":44.9,"maker":"moonshot"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":45.8,"maker":"google"},{"date":"2026-01-27","model":"Kimi K2.5","score":51.8,"maker":"moonshot"},{"date":"2026-04-07","model":"Claude Mythos Preview","score":64.7,"maker":"anthropic"},{"date":"2026-09-01","model":"Fable 5.1","score":65.6,"maker":"anthropic"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":67.7,"maker":"anthropic"},{"date":"2026-10-23","model":"Kimi K3.1 Pro","score":70,"maker":"moonshot"}],"models":[{"model":"Kimi K3.1 Pro","score":70,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":70,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":1},{"model":"Claude Opus 5.5","score":67.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Opus 5.5","slug":"claude-opus-55","url":"https://handyai.news/modeldrop/claude-opus-55","published":"2026-09-22"},"launch_score":67.7,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2},{"model":"Fable 5.1","score":65.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":3},{"model":"Claude Mythos Preview","score":64.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":4},{"model":"Mythos 5","score":64.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":64.5,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":5},{"model":"GPT-5.6 Sol","score":64.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":6},{"model":"Fable 5","score":63.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":7},{"model":"Claude Opus 5","score":63.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 5.5","slug":"claude-opus-55","url":"https://handyai.news/modeldrop/claude-opus-55","published":"2026-09-22"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":8},{"model":"GLM-5.3","score":62.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":62.5,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":9},{"model":"Kimi K3","score":59.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":10},{"model":"DeepSeek V4.1 Flash 1017","score":58.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":11},{"model":"Claude Opus 4.8","score":57.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-48","rank":12},{"model":"GPT-6 Astra","score":57.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":13},{"model":"Qwen3.8-Max","score":56.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":56.2,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":14},{"model":"GLM-5.2","score":54.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"first_cited","drop_url":null,"rank":15},{"model":"Kimi K2.6","score":54,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-04-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k26","rank":16},{"model":"GPT-5.5","score":52.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-04-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-55","rank":17},{"model":"Kimi K2.5","score":51.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/modeldrop/glm-5","published":"2026-02-11"},"launch_score":50.2,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":18},{"model":"Gemini 3.1 Pro","score":51.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":19},{"model":"GLM-5","score":50.4,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/modeldrop/glm-5","published":"2026-02-11"},"launch_score":50.4,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-02-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-5","rank":20},{"model":"Inkling","score":46,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":46,"reports":1,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":21},{"model":"Gemini 3 Pro","score":45.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":22},{"model":"GPT-5.2","score":45.5,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":23},{"model":"Kimi K2 Thinking","score":44.9,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":44.9,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-11-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2-thinking","rank":24},{"model":"Claude Opus 4.5","score":43.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":25},{"model":"GPT-5","score":41.7,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":26},{"model":"Grok 4","score":41,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":38.6,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":27},{"model":"DeepSeek V3.2","score":40.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/modeldrop/glm-5","published":"2026-02-11"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-12-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v32","rank":28},{"model":"Nemotron 3 Ultra","score":37.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":29},{"model":"Claude Sonnet 4.5","score":32,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":30},{"model":"MiniMax-M2","score":31.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":31.8,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m2","rank":31},{"model":"GLM-4.6","score":30.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":32},{"model":"DeepSeek V3.1","score":29.8,"own_drop":true,"independent":false,"setting":"thinking","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":29.8,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-08-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v31","rank":33},{"model":"o3","score":24.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":34},{"model":"DeepSeek R1 0528","score":24.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1-0528","rank":35},{"model":"DeepSeek V3.2-Exp","score":20.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":36},{"model":"gpt-oss-120b","score":19,"own_drop":true,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":19,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":37},{"model":"o4-mini","score":17.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":38},{"model":"gpt-oss-20b","score":17.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":39},{"model":"LFM3-1.7B","score":13.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Liquid AI, \"Introducing LFM3\"","slug":"lfm3-17b","url":"https://www.liquid.ai/blog/lfm3","published":"2026-10-27"},"launch_score":13.9,"reports":1,"maker":{"key":"liquid","name":"Liquid AI","logo_url":"https://handyai.news/ui/makers/liquid.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":40}]},{"key":"aime-2024","name":"AIME 2024","about":null,"type":{"key":"math","label":"Math"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#aime-2024","api_url":"https://handyai.news/api/benchmarks/aime-2024","image_url":"https://handyai.news/api/benchmarks/aime-2024.png","leader":{"model":"Grok 4","score":94.3,"maker":"SpaceXAI"},"model_count":23,"drop_count":12,"first_date":"2024-12-05","last_date":"2025-07-28","frontier_gain":20,"frontier":[{"date":"2024-12-05","model":"o1","score":74.3,"maker":"openai"},{"date":"2025-01-20","model":"DeepSeek R1","score":79.8,"maker":"deepseek"},{"date":"2025-01-31","model":"o3-mini","score":87.3,"maker":"openai"},{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":88.7,"maker":"google"},{"date":"2025-04-16","model":"o3","score":90.3,"maker":"openai"},{"date":"2025-06-10","model":"o3-pro","score":93,"maker":"openai"},{"date":"2025-07-09","model":"Grok 4","score":94.3,"maker":"xai"}],"models":[{"model":"Grok 4","score":94.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":1},{"model":"o3-pro","score":93,"own_drop":true,"independent":false,"setting":"","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":93,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-06-10","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-pro","rank":2},{"model":"GLM-4.5","score":91,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":91,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-07-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-45","rank":3},{"model":"o3","score":90.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":4},{"model":"DeepSeek R1 0528","score":89.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":91.4,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1-0528","rank":5},{"model":"Gemini 2.5 Pro","score":88.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":null,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":6},{"model":"o3-mini","score":87.3,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":87.3,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":7},{"model":"o1-pro","score":86,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"first_cited","drop_url":null,"rank":8},{"model":"Qwen3","score":85.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":85.7,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-04-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3","rank":9},{"model":"Qwen3 235B","score":85.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"first_cited","drop_url":null,"rank":10},{"model":"Grok 3","score":83.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 3","slug":"grok-3","url":"https://handyai.news/modeldrop/grok-3","published":"2025-02-17"},"launch_score":83.9,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-3","rank":11},{"model":"Grok 3 Beta","score":83.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"first_cited","drop_url":null,"rank":12},{"model":"Llama 3.1 Nemotron Ultra","score":80.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Llama 3.1 Nemotron Ultra","slug":"llama-31-nemotron-ultra","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","published":"2025-04-07"},"launch_score":80.8,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","rank":13},{"model":"DeepSeek R1","score":79.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":79.8,"reports":6,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":14},{"model":"o1","score":74.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":79.2,"reports":6,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":15},{"model":"o1-mini","score":63.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-20","dated_by":"first_cited","drop_url":null,"rank":16},{"model":"Claude 3.7 Sonnet","score":61.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude 3.7 Sonnet","slug":"claude-37-sonnet","url":"https://handyai.news/modeldrop/claude-37-sonnet","published":"2025-02-24"},"launch_score":61.3,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":17},{"model":"o1-preview","score":42,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o1","slug":"o1","url":"https://handyai.news/modeldrop/o1","published":"2024-12-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":18},{"model":"DeepSeek V3","score":39.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":39.2,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v3","rank":19},{"model":"GPT-4.5","score":36.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":36.7,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":20},{"model":"Llama 3.1 405B","score":23.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"first_cited","drop_url":null,"rank":21},{"model":"Claude 3.5 Sonnet","score":16,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude 3.7 Sonnet","slug":"claude-37-sonnet","url":"https://handyai.news/modeldrop/claude-37-sonnet","published":"2025-02-24"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":22},{"model":"GPT-4o","score":9.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":23}]},{"key":"bench-telecom","name":"τ²-bench Telecom","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#bench-telecom","api_url":"https://handyai.news/api/benchmarks/bench-telecom","image_url":"https://handyai.news/api/benchmarks/bench-telecom.png","leader":{"model":"GPT-5.2","score":98.7,"maker":"OpenAI"},"model_count":26,"drop_count":11,"first_date":"2025-04-14","last_date":"2026-10-15","frontier_gain":64.7,"frontier":[{"date":"2025-04-14","model":"GPT-4.1","score":34,"maker":"openai"},{"date":"2025-04-16","model":"o3","score":58.2,"maker":"openai"},{"date":"2025-07-11","model":"Kimi K2","score":65.8,"maker":"moonshot"},{"date":"2025-08-05","model":"Claude Opus 4.1","score":71.5,"maker":"anthropic"},{"date":"2025-08-07","model":"GPT-5","score":96.7,"maker":"openai"},{"date":"2025-09-29","model":"Claude Sonnet 4.5","score":98,"maker":"anthropic"},{"date":"2025-12-11","model":"GPT-5.2","score":98.7,"maker":"openai"}],"models":[{"model":"GPT-5.2","score":98.7,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.2","slug":"gpt-52","url":"https://handyai.news/modeldrop/gpt-52","published":"2025-12-11"},"launch_score":98.7,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":1},{"model":"GLM-5.3-Flash","score":98.46,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"first_cited","drop_url":null,"rank":2},{"model":"Claude Sonnet 4.5","score":98,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":98,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":3},{"model":"GPT-5","score":96.7,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.1","slug":"gpt-51","url":"https://handyai.news/modeldrop/gpt-51","published":"2025-11-12"},"launch_score":96.7,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":4},{"model":"GPT-5.1","score":95.6,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"GPT-5.1","slug":"gpt-51","url":"https://handyai.news/modeldrop/gpt-51","published":"2025-11-12"},"launch_score":95.6,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":5},{"model":"GPT-5.1 Thinking","score":95.6,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.2","slug":"gpt-52","url":"https://handyai.news/modeldrop/gpt-52","published":"2025-12-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"first_cited","drop_url":null,"rank":6},{"model":"Kimi K2 Thinking","score":93,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":93,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-11-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2-thinking","rank":7},{"model":"Qwen3.5-122B-A10B","score":87.17,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"first_cited","drop_url":null,"rank":8},{"model":"MiniMax-M2","score":86.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":null,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m2","rank":9},{"model":"Nemotron 3 Ultra","score":83,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":10},{"model":"Nemotron 3.5 Super VL","score":73.19,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":73.19,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":11},{"model":"MAI-Code-1-Flash","score":71.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":71.7,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-06-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-code-1-flash","rank":12},{"model":"Claude Opus 4.1","score":71.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":13},{"model":"gpt-oss-120b","score":66,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":14},{"model":"Kimi K2","score":65.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":65.8,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-07-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2","rank":15},{"model":"Nemotron 3 Super","score":64.36,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":64.36,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-super","rank":16},{"model":"o3","score":58.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":17},{"model":"Claude Opus 4","score":57,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":18},{"model":"Claude Haiku 4.5","score":54.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":83,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":19},{"model":"gpt-oss-20b","score":49.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":20},{"model":"Claude Sonnet 4","score":49.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":21},{"model":"Nemotron 3 Nano","score":42.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":42.2,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":22},{"model":"GPT-4.1","score":34,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":23},{"model":"DeepSeek V3.2-Exp","score":33.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":24},{"model":"DeepSeek V3 0324","score":32.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-14","dated_by":"first_cited","drop_url":null,"rank":25},{"model":"Qwen3-30B-A3B-Thinking-2507","score":26.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"first_cited","drop_url":null,"rank":26}]},{"key":"simpleqa","name":"SimpleQA","about":null,"type":{"key":"knowledge","label":"Knowledge"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#simpleqa","api_url":"https://handyai.news/api/benchmarks/simpleqa","image_url":"https://handyai.news/api/benchmarks/simpleqa.png","leader":{"model":"Grok 4 Fast","score":95,"maker":"SpaceXAI"},"model_count":23,"drop_count":11,"first_date":"2024-12-05","last_date":"2025-09-19","frontier_gain":48,"frontier":[{"date":"2024-12-05","model":"o1","score":47,"maker":"openai"},{"date":"2025-02-17","model":"Grok 3","score":82,"maker":"xai"},{"date":"2025-07-09","model":"Grok 4","score":94,"maker":"xai"},{"date":"2025-09-19","model":"Grok 4 Fast","score":95,"maker":"xai"}],"models":[{"model":"Grok 4 Fast","score":95,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":95,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-09-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4-fast","rank":1},{"model":"Grok 4","score":94,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":2},{"model":"Grok 3","score":82,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":43.6,"reports":3,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-3","rank":3},{"model":"GPT-4.5","score":62.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Pro","slug":"gemini-25-pro","url":"https://handyai.news/modeldrop/gemini-25-pro","published":"2025-03-25"},"launch_score":62.5,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":4},{"model":"Gemini 2.5 Pro","score":52.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Pro","slug":"gemini-25-pro","url":"https://handyai.news/modeldrop/gemini-25-pro","published":"2025-03-25"},"launch_score":52.9,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":5},{"model":"o1","score":47,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":47,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":6},{"model":"Gemini 2.0 Pro","score":44.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 3","slug":"grok-3","url":"https://handyai.news/modeldrop/grok-3","published":"2025-02-17"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-02-17","dated_by":"first_cited","drop_url":null,"rank":7},{"model":"Grok 3 Beta","score":43.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"first_cited","drop_url":null,"rank":8},{"model":"GPT-4.1","score":42.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":9},{"model":"o1-preview","score":42,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o1","slug":"o1","url":"https://handyai.news/modeldrop/o1","published":"2024-12-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":10},{"model":"GPT-4o","score":38.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":11},{"model":"Kimi K2","score":31,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":31,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-07-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2","rank":12},{"model":"DeepSeek R1","score":30.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":30.1,"reports":4,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":13},{"model":"Gemini 2.0 Flash","score":29.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":14},{"model":"Gemini 2.5 Flash","score":29.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":29.7,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-flash","rank":15},{"model":"Claude 3.5 Sonnet","score":28.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":16},{"model":"DeepSeek R1 0528","score":27.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek R1 0528","slug":"deepseek-r1-0528","url":"https://handyai.news/modeldrop/deepseek-r1-0528","published":"2025-05-28"},"launch_score":27.8,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1-0528","rank":17},{"model":"DeepSeek V3 0324","score":27.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-14","dated_by":"first_cited","drop_url":null,"rank":18},{"model":"DeepSeek V3","score":24.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 3","slug":"grok-3","url":"https://handyai.news/modeldrop/grok-3","published":"2025-02-17"},"launch_score":24.9,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v3","rank":19},{"model":"Claude Opus 4","score":22.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":20},{"model":"Llama 3.1 405B","score":17.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"first_cited","drop_url":null,"rank":21},{"model":"o3-mini","score":13.8,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Gemini 2.5 Pro","slug":"gemini-25-pro","url":"https://handyai.news/modeldrop/gemini-25-pro","published":"2025-03-25"},"launch_score":13.8,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":22},{"model":"o1-mini","score":7.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-20","dated_by":"first_cited","drop_url":null,"rank":23}]},{"key":"osworld-verified","name":"OSWorld-Verified","about":"Using a computer: apps, files and the web (2025 edition).","type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#osworld-verified","api_url":"https://handyai.news/api/benchmarks/osworld-verified","image_url":"https://handyai.news/api/benchmarks/osworld-verified.png","leader":{"model":"Kimi K3","score":84.8,"maker":"Moonshot AI"},"model_count":24,"drop_count":10,"first_date":"2025-05-22","last_date":"2026-09-21","frontier_gain":42.6,"frontier":[{"date":"2025-05-22","model":"Claude Sonnet 4","score":42.2,"maker":"anthropic"},{"date":"2025-08-05","model":"Claude Opus 4.1","score":44.4,"maker":"anthropic"},{"date":"2025-09-29","model":"Claude Sonnet 4.5","score":61.4,"maker":"anthropic"},{"date":"2025-11-24","model":"Claude Opus 4.5","score":66.3,"maker":"anthropic"},{"date":"2026-02-05","model":"GPT-5.3-Codex","score":74,"maker":"openai"},{"date":"2026-03-05","model":"GPT-5.4","score":75,"maker":"openai"},{"date":"2026-04-07","model":"Claude Mythos Preview","score":79.6,"maker":"anthropic"},{"date":"2026-06-26","model":"GPT-5.6 Sol","score":83,"maker":"openai"},{"date":"2026-07-16","model":"Kimi K3","score":84.8,"maker":"moonshot"}],"models":[{"model":"Kimi K3","score":84.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":1},{"model":"Qwen3.8","score":84.3,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":84.3,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":2},{"model":"Claude Opus 5","score":83.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":3},{"model":"Gemini 3.6 Flash","score":83,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":83,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":4},{"model":"GPT-5.6 Sol","score":83,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":5},{"model":"MiMo-V2.6-Pro","score":82,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":82,"reports":1,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":6},{"model":"Claude Sonnet 5","score":81.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":7},{"model":"Claude Mythos Preview","score":79.6,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Claude Mythos Preview","slug":"claude-mythos-preview","url":"https://handyai.news/modeldrop/claude-mythos-preview","published":"2026-04-07"},"launch_score":79.6,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":8},{"model":"Gemini 3.5 Flash","score":78.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-05-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-35-flash","rank":9},{"model":"GPT-5.4","score":75,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Mythos Preview","slug":"claude-mythos-preview","url":"https://handyai.news/modeldrop/claude-mythos-preview","published":"2026-04-07"},"launch_score":75,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":10},{"model":"GPT-5.3-Codex","score":74,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":64.7,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-53-codex","rank":11},{"model":"Qwen3.7-Plus","score":73.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":12},{"model":"Claude Opus 4.6","score":72.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":72.7,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":13},{"model":"GPT-5.6 Luna","score":72.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":14},{"model":"Claude Sonnet 4.6","score":72.5,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":72.5,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-46","rank":15},{"model":"Claude Opus 4.5","score":66.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":66.3,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":16},{"model":"Muse Glimmer-30B","score":65.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":17},{"model":"Qwen3.6-27B","score":63.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8","slug":"qwen38","url":"https://handyai.news/modeldrop/qwen38","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":18},{"model":"Claude Sonnet 4.5","score":61.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.6","slug":"claude-sonnet-46","url":"https://handyai.news/modeldrop/claude-sonnet-46","published":"2026-02-17"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":19},{"model":"Claude Haiku 4.5","score":50.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":50.7,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":20},{"model":"GPT-5.2","score":47.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":21},{"model":"Claude Opus 4.1","score":44.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.5","slug":"claude-opus-45","url":"https://handyai.news/modeldrop/claude-opus-45","published":"2025-11-24"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":22},{"model":"Claude Sonnet 4","score":42.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":23},{"model":"GPT-5.2-Codex","score":38.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.3-Codex","slug":"gpt-53-codex","url":"https://handyai.news/modeldrop/gpt-53-codex","published":"2026-02-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52-codex","rank":24}]},{"key":"mmlu-pro","name":"MMLU-Pro","about":null,"type":{"key":"knowledge","label":"Knowledge"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#mmlu-pro","api_url":"https://handyai.news/api/benchmarks/mmlu-pro","image_url":"https://handyai.news/api/benchmarks/mmlu-pro.png","leader":{"model":"Claude Opus 4","score":87.3,"maker":"Anthropic"},"model_count":25,"drop_count":9,"first_date":"2024-12-05","last_date":"2026-10-28","frontier_gain":9.3,"frontier":[{"date":"2024-12-05","model":"Claude 3.5 Sonnet","score":78,"maker":"anthropic"},{"date":"2025-02-17","model":"Grok 3","score":79.9,"maker":"xai"},{"date":"2025-02-24","model":"Claude 3.7 Sonnet","score":80,"maker":"anthropic"},{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":86.2,"maker":"google"},{"date":"2025-05-22","model":"Claude Opus 4","score":87.3,"maker":"anthropic"}],"models":[{"model":"Claude Opus 4","score":87.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":1},{"model":"Claude Sonnet 4.6","score":87,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-46","rank":2},{"model":"Gemini 2.5 Pro","score":86.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":3},{"model":"o3","score":85.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":4},{"model":"Gemma 4","score":85.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":85.2,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":5},{"model":"MAI-Thinking-1","score":85,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Thinking-1","slug":"mai-thinking-1","url":"https://handyai.news/modeldrop/mai-thinking-1","published":"2026-08-12"},"launch_score":85,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-thinking-1","rank":6},{"model":"DeepSeek R1 0528","score":84.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1-0528","rank":7},{"model":"Clover-35B-A3B-Preview","score":84.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DiffuSpace, \"Clover-35B-A3B-Preview\"","slug":"clover-35b-a3b-preview","url":"https://diffuspace.ai/blog/clover-35b/","published":"2026-10-28"},"launch_score":84.8,"reports":1,"maker":{"key":"diffuspace","name":"DiffuSpace","logo_url":"https://handyai.news/ui/makers/diffuspace.svg"},"open_weights":true,"released":"2026-10-28","dated_by":"own_drop","drop_url":null,"rank":8},{"model":"GLM-4.5","score":84.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":84.6,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-07-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-45","rank":9},{"model":"Gemma 4 26B A4B","score":82.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":10},{"model":"DeepSeek V3-0324","score":81.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"first_cited","drop_url":null,"rank":11},{"model":"Llama 4 Maverick","score":80.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":80.5,"reports":3,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":12},{"model":"Claude 3.7 Sonnet","score":80,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":13},{"model":"Grok 3","score":79.9,"own_drop":true,"independent":false,"setting":"no thinking","source":{"drop":"Grok 3","slug":"grok-3","url":"https://handyai.news/modeldrop/grok-3","published":"2025-02-17"},"launch_score":79.9,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-3","rank":14},{"model":"Gemini 2.0 Pro","score":79.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 3","slug":"grok-3","url":"https://handyai.news/modeldrop/grok-3","published":"2025-02-17"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-02-17","dated_by":"first_cited","drop_url":null,"rank":15},{"model":"Claude 3.5 Sonnet","score":78,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 3","slug":"grok-3","url":"https://handyai.news/modeldrop/grok-3","published":"2025-02-17"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":16},{"model":"Gemini 2.0 Flash","score":77.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":76.4,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":17},{"model":"Mistral Medium 3","score":77.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":77.2,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2025-05-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-medium-3","rank":18},{"model":"DeepSeek V3","score":75.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 3","slug":"grok-3","url":"https://handyai.news/modeldrop/grok-3","published":"2025-02-17"},"launch_score":75.9,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v3","rank":19},{"model":"Gemini 1.5 Pro","score":75.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":20},{"model":"GPT-4o","score":75.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":21},{"model":"Llama 4 Scout","score":74.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":74.3,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":22},{"model":"Llama 3.1 405B","score":73.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"first_cited","drop_url":null,"rank":23},{"model":"Gemma 3 27B","score":67.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"first_cited","drop_url":null,"rank":24},{"model":"Gemini 1.5 Flash","score":67.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"first_cited","drop_url":null,"rank":25}]},{"key":"ifbench","name":"IFBench","about":null,"type":{"key":"instructions","label":"Instructions"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#ifbench","api_url":"https://handyai.news/api/benchmarks/ifbench","image_url":"https://handyai.news/api/benchmarks/ifbench.png","leader":{"model":"Nemotron 3 Ultra","score":82.94,"maker":"NVIDIA"},"model_count":24,"drop_count":8,"first_date":"2025-08-05","last_date":"2026-10-27","frontier_gain":14.6,"frontier":[{"date":"2025-08-05","model":"gpt-oss-120b","score":68.32,"maker":"openai"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":70.4,"maker":"google"},{"date":"2025-12-11","model":"GPT-5.2","score":75.4,"maker":"openai"},{"date":"2026-02-16","model":"Qwen3.5","score":76.5,"maker":"qwen"},{"date":"2026-03-27","model":"GLM-5.1","score":76.6,"maker":"zai"},{"date":"2026-06-04","model":"Nemotron 3 Ultra","score":82.94,"maker":"nvidia"}],"models":[{"model":"Nemotron 3 Ultra","score":82.94,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":81.7,"reports":3,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":1},{"model":"Qwen3.8-Max","score":82.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":82.8,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":2},{"model":"Inkling","score":79.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":79.8,"reports":1,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":3},{"model":"Qwen3.5 397B","score":78.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"first_cited","drop_url":null,"rank":4},{"model":"GLM-5.1","score":76.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-03-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-51","rank":5},{"model":"Qwen3.5","score":76.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":76.5,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-02-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen35","rank":6},{"model":"Kimi K2.6","score":76,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-04-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k26","rank":7},{"model":"GPT-5.2","score":75.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":8},{"model":"GLM-5.2","score":73.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"first_cited","drop_url":null,"rank":9},{"model":"LFM3-1.7B","score":73.03,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Liquid AI, \"Introducing LFM3\"","slug":"lfm3-17b","url":"https://www.liquid.ai/blog/lfm3","published":"2026-10-27"},"launch_score":73.03,"reports":1,"maker":{"key":"liquid","name":"Liquid AI","logo_url":"https://handyai.news/ui/makers/liquid.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":10},{"model":"GPT-5.6 Sol","score":72.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":11},{"model":"Nemotron 3 Super","score":72.56,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":72.56,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-super","rank":12},{"model":"Nemotron 3 Nano","score":71.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":71.5,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":13},{"model":"Nemotron 3.5 Super VL","score":70.86,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":70.86,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":14},{"model":"Gemini 3 Pro","score":70.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":15},{"model":"Kimi K2.5","score":70.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":16},{"model":"gpt-oss-120b","score":68.32,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":17},{"model":"Qwen3.5-122B-A10B","score":67.56,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"first_cited","drop_url":null,"rank":18},{"model":"gpt-oss-20b","score":65,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":19},{"model":"Fable 5","score":63.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":20},{"model":"Claude Opus 4.8","score":62.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-48","rank":21},{"model":"Claude Opus 4.5","score":58,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":22},{"model":"GLM-5.3-Flash","score":57.67,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"first_cited","drop_url":null,"rank":23},{"model":"Qwen3-30B-A3B-Thinking-2507","score":51,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"first_cited","drop_url":null,"rank":24}]},{"key":"lmarena-text-elo","name":"LMArena text Elo","about":null,"type":{"key":"overall","label":"Overall index","folded":true},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#lmarena-text-elo","api_url":"https://handyai.news/api/benchmarks/lmarena-text-elo","image_url":"https://handyai.news/api/benchmarks/lmarena-text-elo.png","leader":{"model":"Grok 4.1","score":1483,"maker":"SpaceXAI"},"model_count":23,"drop_count":8,"first_date":"2024-12-05","last_date":"2025-12-17","frontier_gain":106,"frontier":[{"date":"2024-12-05","model":"GPT-4o","score":1377,"maker":"openai"},{"date":"2025-01-31","model":"Gemini 2.0 Flash Thinking","score":1385,"maker":"google"},{"date":"2025-02-17","model":"Grok 3","score":1404,"maker":"xai"},{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":1452,"maker":"google"},{"date":"2025-11-17","model":"Grok 4.1","score":1483,"maker":"xai"}],"models":[{"model":"Grok 4.1","score":1483,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":1483,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-11-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-41","rank":1},{"model":"Gemini 3 Flash","score":1470,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":1470,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-flash","rank":2},{"model":"Gemini 2.5 Pro","score":1452,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":1443,"reports":4,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":3},{"model":"GPT-5.1","score":1441,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":4},{"model":"o3","score":1427,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":5},{"model":"Grok 4 Fast","score":1425,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":1425,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-09-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4-fast","rank":6},{"model":"Grok 4","score":1422,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":7},{"model":"Claude Opus 4.1","score":1419,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":1419,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":8},{"model":"Claude Sonnet 4.5","score":1419,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":1419,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":9},{"model":"GPT-5.2","score":1419,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":10},{"model":"GPT-5","score":1410,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":11},{"model":"Grok 4.1 Fast","score":1410,"own_drop":false,"independent":true,"setting":"reasoning","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"first_cited","drop_url":null,"rank":12},{"model":"Gemini 2.5 Flash","score":1407,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-flash","rank":13},{"model":"Grok 3","score":1404,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 2.5 Pro","slug":"gemini-25-pro","url":"https://handyai.news/modeldrop/gemini-25-pro","published":"2025-03-25"},"launch_score":1402,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-3","rank":14},{"model":"GPT-4.5","score":1398,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 2.5 Pro","slug":"gemini-25-pro","url":"https://handyai.news/modeldrop/gemini-25-pro","published":"2025-03-25"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":15},{"model":"GPT-5 mini","score":1393,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-28","dated_by":"first_cited","drop_url":null,"rank":16},{"model":"Gemini 2.0 Flash Thinking","score":1385,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 3","slug":"grok-3","url":"https://handyai.news/modeldrop/grok-3","published":"2025-02-17"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"first_cited","drop_url":null,"rank":17},{"model":"Claude Haiku 4.5","score":1384,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":18},{"model":"GPT-4o","score":1377,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 3","slug":"grok-3","url":"https://handyai.news/modeldrop/grok-3","published":"2025-02-17"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":19},{"model":"Gemini 2.0 Flash","score":1355,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":1355,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":20},{"model":"Claude 3.7 Sonnet","score":1304,"own_drop":false,"independent":true,"setting":"thinking","source":{"drop":"Gemini 2.5 Pro","slug":"gemini-25-pro","url":"https://handyai.news/modeldrop/gemini-25-pro","published":"2025-03-25"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":21},{"model":"GPT-4o mini","score":1273,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"first_cited","drop_url":null,"rank":22},{"model":"Claude 3.5 Haiku","score":1238,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"first_cited","drop_url":null,"rank":23}]},{"key":"terminal-bench","name":"Terminal-Bench","about":"Tasks in a real terminal (first edition).","type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#terminal-bench","api_url":"https://handyai.news/api/benchmarks/terminal-bench","image_url":"https://handyai.news/api/benchmarks/terminal-bench.png","leader":{"model":"Claude Sonnet 4.5","score":51,"maker":"Anthropic"},"model_count":21,"drop_count":8,"first_date":"2025-02-24","last_date":"2025-11-06","frontier_gain":15.8,"frontier":[{"date":"2025-02-24","model":"Claude 3.7 Sonnet","score":35.2,"maker":"anthropic"},{"date":"2025-05-22","model":"Claude Opus 4","score":39.2,"maker":"anthropic"},{"date":"2025-08-05","model":"Claude Opus 4.1","score":46.5,"maker":"anthropic"},{"date":"2025-09-29","model":"Claude Sonnet 4.5","score":51,"maker":"anthropic"}],"models":[{"model":"Claude Sonnet 4.5","score":51,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":50,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":1},{"model":"Kimi K2 Thinking","score":47.1,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":47.1,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-11-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2-thinking","rank":2},{"model":"Claude Opus 4.1","score":46.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":43.3,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":3},{"model":"MiniMax-M2","score":46.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":46.3,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m2","rank":4},{"model":"Kimi K2 0905","score":44.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-09-24","dated_by":"first_cited","drop_url":null,"rank":5},{"model":"GPT-5","score":43.8,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":6},{"model":"Claude Haiku 4.5","score":41,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":41,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":7},{"model":"GLM-4.6","score":40.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":8},{"model":"Claude Opus 4","score":39.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":43.2,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":9},{"model":"DeepSeek V3.2-Exp","score":37.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2 Thinking","slug":"kimi-k2-thinking","url":"https://handyai.news/modeldrop/kimi-k2-thinking","published":"2025-11-06"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":10},{"model":"GLM-4.5","score":37.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":37.5,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-07-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-45","rank":11},{"model":"Claude Sonnet 4","score":36.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":null,"reports":5,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":12},{"model":"Claude 3.7 Sonnet","score":35.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":13},{"model":"DeepSeek V3.1","score":31.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":31.3,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-08-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v31","rank":14},{"model":"GPT-4.1","score":30.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":15},{"model":"o3","score":30.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":16},{"model":"Gemini 2.5 Pro Preview","score":25.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"first_cited","drop_url":null,"rank":17},{"model":"Gemini 2.5 Pro","score":25.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 4.5","slug":"claude-haiku-45","url":"https://handyai.news/modeldrop/claude-haiku-45","published":"2025-10-15"},"launch_score":null,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":18},{"model":"Kimi K2","score":25,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-07-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2","rank":19},{"model":"DeepSeek V3 0324","score":13.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-14","dated_by":"first_cited","drop_url":null,"rank":20},{"model":"DeepSeek R1 0528","score":5.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3.1","slug":"deepseek-v31","url":"https://handyai.news/modeldrop/deepseek-v31","published":"2025-08-21"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1-0528","rank":21}]},{"key":"osworld-2","name":"OSWorld 2","about":"Using a computer: apps, files and the web.","type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#osworld-2","api_url":"https://handyai.news/api/benchmarks/osworld-2","image_url":"https://handyai.news/api/benchmarks/osworld-2.png","leader":{"model":"Claude Sonnet 5.5","score":83.9,"maker":"Anthropic"},"model_count":16,"drop_count":8,"first_date":"2026-06-26","last_date":"2026-11-04","frontier_gain":18.2,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":65.7,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":70.6,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":72.6,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":81.8,"maker":"anthropic"},{"date":"2026-09-28","model":"Claude Sonnet 5.5","score":83.9,"maker":"anthropic"}],"models":[{"model":"Claude Sonnet 5.5","score":83.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":80.1,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":1},{"model":"Claude Opus 5.5","score":81.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 5.5","slug":"claude-sonnet-55","url":"https://handyai.news/modeldrop/claude-sonnet-55","published":"2026-09-28"},"launch_score":81.8,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2},{"model":"Hy4","score":78.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Tencent Hunyuan Hy4 launch post","slug":"hy4","url":"https://x.com/TencentHunyuan/status/2117870250838630539","published":"2026-11-04"},"launch_score":78.9,"reports":1,"maker":{"key":"tencent","name":"Tencent","logo_url":"https://handyai.news/ui/makers/tencent.svg"},"open_weights":true,"released":"2026-11-04","dated_by":"own_drop","drop_url":null,"rank":3},{"model":"GPT-6.1 Sol","score":78,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":4},{"model":"GPT-6.1 Terra","score":74.2,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":74.2,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":5},{"model":"GPT-6 Astra","score":72.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-6 Astra","slug":"gpt-6-astra","url":"https://handyai.news/modeldrop/gpt-6-astra","published":"2026-09-03"},"launch_score":72.6,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":6},{"model":"Claude Opus 5","score":70.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":7},{"model":"Muse Spark 1.3","score":66.9,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":66.9,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-13","rank":8},{"model":"GPT-5.6 Sol","score":65.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":9},{"model":"MiniMax M3.1 Flash","score":61.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":61.7,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":10},{"model":"Gemini 3.8 Flash","score":59,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":59,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":11},{"model":"Kimi K3","score":58.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":12},{"model":"GPT-5.6 Terra","score":57.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":13},{"model":"Claude Sonnet 5","score":57,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 5.5","slug":"claude-sonnet-55","url":"https://handyai.news/modeldrop/claude-sonnet-55","published":"2026-09-28"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":14},{"model":"GPT-6 Luna","score":55.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":15},{"model":"DeepSeek V4.1 Flash","score":54.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":16}]},{"key":"harvey-legal-agent-benchmark","name":"Harvey Legal Agent Benchmark","about":"Legal work end to end.","type":{"key":"work","label":"Work"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#harvey-legal-agent-benchmark","api_url":"https://handyai.news/api/benchmarks/harvey-legal-agent-benchmark","image_url":"https://handyai.news/api/benchmarks/harvey-legal-agent-benchmark.png","leader":{"model":"Kimi K3.1 Pro","score":23.7,"maker":"Moonshot AI"},"model_count":21,"drop_count":7,"first_date":"2026-06-09","last_date":"2026-10-23","frontier_gain":12.4,"frontier":[{"date":"2026-06-09","model":"Fable 5","score":11.3,"maker":"anthropic"},{"date":"2026-07-09","model":"Grok 4.5","score":12.9,"maker":"xai"},{"date":"2026-08-12","model":"Grok 4.6","score":15.8,"maker":"xai"},{"date":"2026-09-01","model":"Fable 5.1","score":19.1,"maker":"anthropic"},{"date":"2026-09-21","model":"Grok 4.7","score":19.6,"maker":"xai"},{"date":"2026-10-17","model":"DeepSeek V4.1 Flash 1017","score":21,"maker":"deepseek"},{"date":"2026-10-23","model":"Kimi K3.1 Pro","score":23.7,"maker":"moonshot"}],"models":[{"model":"Kimi K3.1 Pro","score":23.7,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":23.7,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":1},{"model":"DeepSeek V4.1 Flash 1017","score":21,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":2},{"model":"Grok 4.7","score":19.6,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":19.6,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":3},{"model":"Gemini 4 Argon","score":19.6,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Argon","slug":"gemini-4-argon","url":"https://handyai.news/modeldrop/gemini-4-argon","published":"2026-10-01"},"launch_score":19.6,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":4},{"model":"Fable 5.1","score":19.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":5},{"model":"Grok 4.6","score":15.8,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":15.8,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":6},{"model":"Mistral Large 4","score":15.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":15.8,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":7},{"model":"Muse Wonder Full","score":13.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":8},{"model":"Grok 4.5","score":12.9,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":9},{"model":"Kimi K3","score":12.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":10},{"model":"Fable 5","score":11.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":11},{"model":"Qwen3.8 Max","score":10.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-10-06","dated_by":"first_cited","drop_url":null,"rank":12},{"model":"Gemini 3.8 Flash","score":10,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":10,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":13},{"model":"Muse Wonder","score":10,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":10,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":14},{"model":"GPT-6 Astra","score":8.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":15},{"model":"GLM-5.3","score":8.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":16},{"model":"Claude Opus 5","score":6.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":17},{"model":"Claude Sonnet 5","score":5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":18},{"model":"Claude Opus 5.5","score":4.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":19},{"model":"GPT-5.6 Sol","score":2.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":20},{"model":"GPT-5.6 Terra","score":0.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":21}]},{"key":"mcp-atlas","name":"MCP Atlas","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#mcp-atlas","api_url":"https://handyai.news/api/benchmarks/mcp-atlas","image_url":"https://handyai.news/api/benchmarks/mcp-atlas.png","leader":{"model":"Muse Spark 1.2","score":90.3,"maker":"Meta"},"model_count":17,"drop_count":6,"first_date":"2025-08-05","last_date":"2026-10-27","frontier_gain":49.4,"frontier":[{"date":"2025-08-05","model":"Claude Opus 4.1","score":40.9,"maker":"anthropic"},{"date":"2025-09-29","model":"Claude Sonnet 4.5","score":43.8,"maker":"anthropic"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":66.6,"maker":"google"},{"date":"2025-12-11","model":"GPT-5.2","score":68,"maker":"openai"},{"date":"2026-06-09","model":"Fable 5","score":83.3,"maker":"anthropic"},{"date":"2026-07-24","model":"Claude Opus 5","score":85.8,"maker":"anthropic"},{"date":"2026-08-05","model":"Muse Spark 1.2","score":90.3,"maker":"meta"}],"models":[{"model":"Muse Spark 1.2","score":90.3,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark 1.2","slug":"muse-spark-12","url":"https://handyai.news/modeldrop/muse-spark-12","published":"2026-08-05"},"launch_score":90.3,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-12","rank":1},{"model":"Muse Spark Open","score":89.2,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":89.2,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":2},{"model":"Claude Opus 5","score":85.8,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark 1.2","slug":"muse-spark-12","url":"https://handyai.news/modeldrop/muse-spark-12","published":"2026-08-05"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":3},{"model":"Qwen3.8-2.4T-A95B","score":84.5,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":4},{"model":"GLM-5.3","score":84.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":5},{"model":"Fable 5","score":83.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Muse Spark 1.2","slug":"muse-spark-12","url":"https://handyai.news/modeldrop/muse-spark-12","published":"2026-08-05"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":6},{"model":"Kimi K3","score":82.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":7},{"model":"GPT-5.6 Sol","score":81.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Muse Spark 1.2","slug":"muse-spark-12","url":"https://handyai.news/modeldrop/muse-spark-12","published":"2026-08-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":8},{"model":"GPT-5.2","score":68,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/modeldrop/glm-5","published":"2026-02-11"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":9},{"model":"GLM-5","score":67.8,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/modeldrop/glm-5","published":"2026-02-11"},"launch_score":67.8,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-02-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-5","rank":10},{"model":"Gemini 3 Pro","score":66.6,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/modeldrop/glm-5","published":"2026-02-11"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":11},{"model":"Kimi K2.5","score":63.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5","slug":"glm-5","url":"https://handyai.news/modeldrop/glm-5","published":"2026-02-11"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":12},{"model":"Claude Opus 4.5","score":62.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.5","slug":"claude-opus-45","url":"https://handyai.news/modeldrop/claude-opus-45","published":"2025-11-24"},"launch_score":62.3,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":13},{"model":"Gemini 3 Flash","score":57.4,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":57.4,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-flash","rank":14},{"model":"Claude Sonnet 4.5","score":43.8,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":15},{"model":"Claude Opus 4.1","score":40.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.5","slug":"claude-opus-45","url":"https://handyai.news/modeldrop/claude-opus-45","published":"2025-11-24"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":16},{"model":"LFM3-1.7B","score":32.57,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Liquid AI, \"Introducing LFM3\"","slug":"lfm3-17b","url":"https://www.liquid.ai/blog/lfm3","published":"2026-10-27"},"launch_score":32.57,"reports":1,"maker":{"key":"liquid","name":"Liquid AI","logo_url":"https://handyai.news/ui/makers/liquid.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":17}]},{"key":"mmmlu","name":"MMMLU","about":null,"type":{"key":"knowledge","label":"Knowledge"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#mmmlu","api_url":"https://handyai.news/api/benchmarks/mmmlu","image_url":"https://handyai.news/api/benchmarks/mmmlu.png","leader":{"model":"Claude Mythos Preview","score":92.7,"maker":"Anthropic"},"model_count":14,"drop_count":6,"first_date":"2024-12-05","last_date":"2026-04-07","frontier_gain":5,"frontier":[{"date":"2024-12-05","model":"o1","score":87.7,"maker":"openai"},{"date":"2025-04-16","model":"o3","score":88.8,"maker":"openai"},{"date":"2025-08-05","model":"Claude Opus 4.1","score":89.5,"maker":"anthropic"},{"date":"2026-02-05","model":"Claude Opus 4.6","score":91.1,"maker":"anthropic"},{"date":"2026-04-07","model":"Claude Mythos Preview","score":92.7,"maker":"anthropic"}],"models":[{"model":"Claude Mythos Preview","score":92.7,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Claude Mythos Preview","slug":"claude-mythos-preview","url":"https://handyai.news/modeldrop/claude-mythos-preview","published":"2026-04-07"},"launch_score":92.7,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":1},{"model":"Claude Opus 4.6","score":91.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Mythos Preview","slug":"claude-mythos-preview","url":"https://handyai.news/modeldrop/claude-mythos-preview","published":"2026-04-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":2},{"model":"Claude Opus 4.1","score":89.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":89.5,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":3},{"model":"Claude Opus 4","score":88.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":88.8,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":4},{"model":"o3","score":88.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":5},{"model":"o1","score":87.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude 3.7 Sonnet","slug":"claude-37-sonnet","url":"https://handyai.news/modeldrop/claude-37-sonnet","published":"2025-02-24"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":6},{"model":"Claude Sonnet 4","score":86.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":7},{"model":"Claude 3.7 Sonnet","score":85.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":86.1,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":8},{"model":"GPT-4.5","score":85.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":85.1,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":9},{"model":"Llama 4 Maverick","score":84.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":84.6,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":10},{"model":"GPT-4.1","score":83.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":11},{"model":"Claude 3.5 Sonnet","score":82.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude 3.7 Sonnet","slug":"claude-37-sonnet","url":"https://handyai.news/modeldrop/claude-37-sonnet","published":"2025-02-24"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":12},{"model":"GPT-4o","score":81.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":13},{"model":"o3-mini","score":81.1,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":14}]},{"key":"tau-bench-retail","name":"TAU-bench Retail","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#tau-bench-retail","api_url":"https://handyai.news/api/benchmarks/tau-bench-retail","image_url":"https://handyai.news/api/benchmarks/tau-bench-retail.png","leader":{"model":"Claude Opus 4.1","score":82.4,"maker":"Anthropic"},"model_count":14,"drop_count":6,"first_date":"2024-12-05","last_date":"2025-08-05","frontier_gain":10.9,"frontier":[{"date":"2024-12-05","model":"Claude 3.5 Sonnet","score":71.5,"maker":"anthropic"},{"date":"2025-02-24","model":"Claude 3.7 Sonnet","score":81.2,"maker":"anthropic"},{"date":"2025-05-22","model":"Claude Opus 4","score":81.4,"maker":"anthropic"},{"date":"2025-08-05","model":"Claude Opus 4.1","score":82.4,"maker":"anthropic"}],"models":[{"model":"Claude Opus 4.1","score":82.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":82.4,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":1},{"model":"Claude Opus 4","score":81.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":81.4,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":2},{"model":"Claude 3.7 Sonnet","score":81.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":81.2,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":3},{"model":"Claude Sonnet 4","score":80.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":4},{"model":"GLM-4.5","score":79.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":79.7,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-07-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-45","rank":5},{"model":"Kimi K2","score":73.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-07-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2","rank":6},{"model":"Claude 3.5 Sonnet","score":71.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude 3.7 Sonnet","slug":"claude-37-sonnet","url":"https://handyai.news/modeldrop/claude-37-sonnet","published":"2025-02-24"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":7},{"model":"o1","score":70.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3 & o4-mini","slug":"o3-o4-mini","url":"https://handyai.news/modeldrop/o3-o4-mini","published":"2025-04-16"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":8},{"model":"o3","score":70.4,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":70.4,"reports":5,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":9},{"model":"GPT-4.1","score":68,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":10},{"model":"gpt-oss-120b","score":67.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":67.8,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":11},{"model":"o4-mini","score":65.6,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":12},{"model":"o3-mini","score":57.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":13},{"model":"gpt-oss-20b","score":54.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":14}]},{"key":"terminal-bench-science","name":"Terminal-Bench Science","about":"Scientific computing tasks in a terminal.","type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#terminal-bench-science","api_url":"https://handyai.news/api/benchmarks/terminal-bench-science","image_url":"https://handyai.news/api/benchmarks/terminal-bench-science.png","leader":{"model":"GPT-6 Astra","score":68.1,"maker":"OpenAI"},"model_count":13,"drop_count":6,"first_date":"2026-06-09","last_date":"2026-11-04","frontier_gain":43.4,"frontier":[{"date":"2026-06-09","model":"Fable 5","score":24.7,"maker":"anthropic"},{"date":"2026-07-24","model":"Claude Opus 5","score":29,"maker":"anthropic"},{"date":"2026-09-01","model":"Fable 5.1","score":40,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":68.1,"maker":"openai"}],"models":[{"model":"GPT-6 Astra","score":68.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":1},{"model":"Claude Opus 5.5","score":63.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":58.7,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2},{"model":"GPT-6.1 Terra","score":58.3,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":58.3,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":3},{"model":"GPT-6.1 Sol","score":57,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":57,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":4},{"model":"Fable 5.1","score":40,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":52.6,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":5},{"model":"Kimi K3.1 Pro","score":38.1,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":38.1,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":6},{"model":"Claude Opus 5","score":29,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 5.5","slug":"claude-opus-55","url":"https://handyai.news/modeldrop/claude-opus-55","published":"2026-09-22"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":7},{"model":"GPT-6 Sol","score":27.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-6.1 Sol","slug":"gpt-61-sol","url":"https://handyai.news/modeldrop/gpt-61-sol","published":"2026-09-29"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":8},{"model":"Fable 5","score":24.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5.1","slug":"fable-51","url":"https://handyai.news/modeldrop/fable-51","published":"2026-09-01"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":9},{"model":"GPT-5.6 Sol","score":22.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5.1","slug":"fable-51","url":"https://handyai.news/modeldrop/fable-51","published":"2026-09-01"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":10},{"model":"Hy4","score":19.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Tencent Hunyuan Hy4 launch post","slug":"hy4","url":"https://x.com/TencentHunyuan/status/2117870250838630539","published":"2026-11-04"},"launch_score":19.8,"reports":1,"maker":{"key":"tencent","name":"Tencent","logo_url":"https://handyai.news/ui/makers/tencent.svg"},"open_weights":true,"released":"2026-11-04","dated_by":"own_drop","drop_url":null,"rank":11},{"model":"DeepSeek V4.1 Flash","score":15.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":12},{"model":"GLM-5.3","score":8.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":13}]},{"key":"epoch-capabilities-index","name":"Epoch Capabilities Index","about":null,"type":{"key":"overall","label":"Overall index","folded":true},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#epoch-capabilities-index","api_url":"https://handyai.news/api/benchmarks/epoch-capabilities-index","image_url":"https://handyai.news/api/benchmarks/epoch-capabilities-index.png","leader":{"model":"GPT-6 Astra","score":169.1,"maker":"OpenAI"},"model_count":93,"drop_count":5,"first_date":"2024-12-05","last_date":"2026-10-26","frontier_gain":27.3,"frontier":[{"date":"2024-12-05","model":"o1","score":141.8,"maker":"openai"},{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":142.2,"maker":"google"},{"date":"2025-04-16","model":"o3","score":146.9,"maker":"openai"},{"date":"2025-06-10","model":"o3-pro","score":147.4,"maker":"openai"},{"date":"2025-08-07","model":"GPT-5","score":150,"maker":"openai"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":153.2,"maker":"google"},{"date":"2025-12-11","model":"GPT-5.2 Pro","score":156.5,"maker":"openai"},{"date":"2026-03-05","model":"GPT-5.4 Pro","score":159.8,"maker":"openai"},{"date":"2026-06-09","model":"Fable 5","score":163,"maker":"anthropic"},{"date":"2026-07-24","model":"Claude Opus 5","score":164.5,"maker":"anthropic"},{"date":"2026-09-01","model":"Fable 5.1","score":166.4,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":169.1,"maker":"openai"}],"models":[{"model":"GPT-6 Astra","score":169.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":1},{"model":"Claude Opus 5.5","score":168.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2},{"model":"GPT-6.1 Terra","score":168.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":3},{"model":"GPT-6.1 Sol","score":167.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":4},{"model":"Claude Sonnet 5.5","score":166.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":5},{"model":"Fable 5.1","score":166.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":6},{"model":"Gemini 4 Argon","score":166.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":7},{"model":"Claude Opus 5","score":164.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":8},{"model":"GPT-6 Sol","score":163.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":9},{"model":"Fable 5","score":163,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":10},{"model":"GPT-5.6 Sol","score":162.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":11},{"model":"Muse Wonder","score":161.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":12},{"model":"Claude Haiku 5.5","score":160.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":13},{"model":"GPT-5.6 Terra","score":160.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":14},{"model":"GPT-5.4 Pro","score":159.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":15},{"model":"Claude Opus 4.8","score":159.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-48","rank":16},{"model":"GPT-5.5","score":159.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-04-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-55","rank":17},{"model":"DeepSeek V4.1 Flash 1017","score":158.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":18},{"model":"Kimi K3","score":158.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":19},{"model":"Gemini 3.7 Flash","score":158.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-08-13","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-37-flash","rank":20},{"model":"Gemini 3.8 Flash","score":157.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":21},{"model":"Muse Spark 1.3","score":157.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":157,"reports":3,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-13","rank":22},{"model":"GPT-5.4","score":157.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":23},{"model":"Claude Sonnet 5","score":157,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":24},{"model":"Grok 4.6","score":156.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":25},{"model":"GPT-6 Luna","score":156.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":26},{"model":"Claude Opus 4.7","score":156.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-47","rank":27},{"model":"GPT-5.6 Luna","score":156.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":28},{"model":"GPT-5.2 Pro","score":156.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":29},{"model":"Qwen3.8-Max","score":156.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":30},{"model":"GLM-5.3","score":156.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":31},{"model":"Claude Opus 4.6","score":155.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":32},{"model":"Gemini 3.1 Pro","score":155.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":33},{"model":"Muse Spark 1.2","score":155.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":155,"reports":3,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-12","rank":34},{"model":"Muse Spark Open","score":155.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":35},{"model":"Gemini 3.5 Flash","score":155,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-05-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-35-flash","rank":36},{"model":"Gemini 3.6 Flash","score":154.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":37},{"model":"Grok 4.7","score":154.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":38},{"model":"Grok 4.5","score":154.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":39},{"model":"Muse Spark 1.1","score":154.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-11","rank":40},{"model":"GPT-5.2","score":153.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":41},{"model":"Gemini 3 Pro","score":153.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":42},{"model":"Claude Sonnet 4.6","score":152.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-46","rank":43},{"model":"Muse Spark","score":152.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":152,"reports":3,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-04-08","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark","rank":44},{"model":"GLM 5.2","score":152.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-06-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-52","rank":45},{"model":"Gemini 3 Flash","score":152,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-flash","rank":46},{"model":"Qwen3.8","score":151.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":47},{"model":"Kimi K2.6","score":151.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-04-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k26","rank":48},{"model":"Mistral Large 4","score":150.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":49},{"model":"Kimi K2.7 Code","score":150.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-06-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k27-code","rank":50},{"model":"Claude Opus 4.5","score":150.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":51},{"model":"GPT-5","score":150,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":52},{"model":"GPT-5.1","score":149.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":53},{"model":"GLM-5.1","score":149.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-03-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-51","rank":54},{"model":"Inkling","score":148.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":55},{"model":"Kimi K2.5","score":148.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":56},{"model":"o3-pro","score":147.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-06-10","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-pro","rank":57},{"model":"o3","score":146.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":58},{"model":"Claude Sonnet 4.5","score":146.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":59},{"model":"Grok 4","score":146.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":60},{"model":"DeepSeek V3.2","score":146.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-12-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v32","rank":61},{"model":"Nemotron 3 Ultra","score":146,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":62},{"model":"Kimi K2 Thinking","score":145.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-11-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2-thinking","rank":63},{"model":"GLM-5","score":145.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-02-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-5","rank":64},{"model":"o4-mini","score":145.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":65},{"model":"Grok 4 Fast","score":144.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-09-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4-fast","rank":66},{"model":"Claude Opus 4.1","score":143.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":67},{"model":"Claude Opus 4","score":142.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":68},{"model":"Qwen3-Max","score":142.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2025-09-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3-max","rank":69},{"model":"Gemini 2.5 Pro","score":142.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (Nov 3)","slug":"epoch-eci-2026-11-03","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-03"},"launch_score":null,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":70},{"model":"Claude Haiku 4.5","score":142.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":71},{"model":"Gemma 4 26B A4B","score":142,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":72},{"model":"o1","score":141.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":73},{"model":"Claude Sonnet 4","score":141.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":74},{"model":"Claude 3.7 Sonnet","score":141,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":75},{"model":"o3-mini","score":140.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":76},{"model":"Gemini 2.5 Flash","score":139.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (Nov 3)","slug":"epoch-eci-2026-11-03","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-03"},"launch_score":null,"reports":4,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-flash","rank":77},{"model":"Kimi K2","score":139.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (Nov 3)","slug":"epoch-eci-2026-11-03","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-03"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-07-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2","rank":78},{"model":"gpt-oss-120b","score":139.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":79},{"model":"DeepSeek V3.1","score":139.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-08-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v31","rank":80},{"model":"DeepSeek R1","score":138.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":81},{"model":"Grok 3","score":138.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-02-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-3","rank":82},{"model":"gpt-oss-20b","score":137.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":83},{"model":"GPT-4.1","score":136.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":84},{"model":"GPT-4.5","score":136.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":85},{"model":"GPT-4.1 mini","score":135,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-05-07","dated_by":"first_cited","drop_url":null,"rank":86},{"model":"Gemini 2.0 Flash","score":134.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (Nov 3)","slug":"epoch-eci-2026-11-03","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-03"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":87},{"model":"Mistral Medium 3","score":134,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":134,"reports":3,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2025-05-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-medium-3","rank":88},{"model":"Llama 4 Maverick","score":132.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":132,"reports":4,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":89},{"model":"DeepSeek V3","score":132.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v3","rank":90},{"model":"DeepSeek V3-0324","score":132,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"first_cited","drop_url":null,"rank":91},{"model":"Llama 4 Scout","score":129.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Epoch Capabilities Index (2026-11-04)","slug":"epoch-auto-2026-11-04","url":"https://epoch.ai/benchmarks/eci","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":92},{"model":"GPT-4o","score":129,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":93}]},{"key":"aa-lcr","name":"AA-LCR","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#aa-lcr","api_url":"https://handyai.news/api/benchmarks/aa-lcr","image_url":"https://handyai.news/api/benchmarks/aa-lcr.png","leader":{"model":"Kimi K3","score":88.7,"maker":"Moonshot AI"},"model_count":37,"drop_count":5,"first_date":"2025-08-05","last_date":"2026-10-28","frontier_gain":55.8,"frontier":[{"date":"2025-08-05","model":"gpt-oss-20b","score":32.88,"maker":"openai"},{"date":"2025-12-15","model":"Qwen3-30B-A3B-Thinking-2507","score":59,"maker":"qwen"},{"date":"2026-04-02","model":"Gemma 4","score":72,"maker":"google"},{"date":"2026-07-15","model":"Inkling","score":77.3,"maker":"thinkingmachines"},{"date":"2026-07-16","model":"Kimi K3","score":88.7,"maker":"moonshot"}],"models":[{"model":"Kimi K3","score":88.7,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":1},{"model":"MiMo-V2.6-Pro","score":86.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":2},{"model":"Kimi K3.1 Pro","score":86.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":3},{"model":"Fable 5.1","score":85.3,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":4},{"model":"Claude Opus 5.5","score":84.7,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":5},{"model":"Nanbeige4.3-33B","score":84,"own_drop":true,"independent":true,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":84,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":6},{"model":"Muse Wonder Full","score":83.7,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":7},{"model":"Muse Glimmer-30B","score":83.3,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":8},{"model":"GPT-6 Luna","score":83.3,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":9},{"model":"GPT-6.1 Sol","score":83,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":10},{"model":"GPT-6.1 Terra","score":82.7,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":11},{"model":"DeepSeek V4.1 Flash 1017","score":82.7,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":12},{"model":"Claude Sonnet 5.5","score":82.7,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":13},{"model":"Claude Haiku 5.5","score":82.7,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":14},{"model":"Ling-3.1-flash-1019","score":82.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"antgroup","name":"Ant Group","logo_url":"https://handyai.news/ui/makers/antgroup.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":15},{"model":"Qwen3.8","score":82,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":16},{"model":"Gemini 3.8 Flash","score":81.3,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":17},{"model":"Muse Wonder","score":80.7,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":18},{"model":"GPT-6 Astra","score":80.7,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":19},{"model":"Qwen3.8-Max","score":80.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":20},{"model":"MiniMax M3.1 Flash","score":79.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":21},{"model":"Gemini 4 Argon","score":79.7,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":22},{"model":"GLM-5.3","score":79.7,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":23},{"model":"Gemini 4 Carbon","score":79.3,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":24},{"model":"Muse Spark Open","score":78.3,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":25},{"model":"Inkling","score":77.3,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":26},{"model":"Grok 4.7","score":76.7,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":27},{"model":"Gemma 4","score":72,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":28},{"model":"Clover-35B-A3B-Preview","score":72,"own_drop":true,"independent":true,"setting":"","source":{"drop":"DiffuSpace, \"Clover-35B-A3B-Preview\"","slug":"clover-35b-a3b-preview","url":"https://diffuspace.ai/blog/clover-35b/","published":"2026-10-28"},"launch_score":72,"reports":1,"maker":{"key":"diffuspace","name":"DiffuSpace","logo_url":"https://handyai.news/ui/makers/diffuspace.svg"},"open_weights":true,"released":"2026-10-28","dated_by":"own_drop","drop_url":null,"rank":29},{"model":"Nemotron 3.5 Super VL","score":69,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":30},{"model":"Qwen3.6 35B A3B","score":61.06,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"first_cited","drop_url":null,"rank":31},{"model":"Qwen3-30B-A3B-Thinking-2507","score":59,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"first_cited","drop_url":null,"rank":32},{"model":"Gemma 4 26B A4B","score":57.56,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":33},{"model":"LFM3-1.7B","score":52.83,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Liquid AI, \"Introducing LFM3\"","slug":"lfm3-17b","url":"https://www.liquid.ai/blog/lfm3","published":"2026-10-27"},"launch_score":52.83,"reports":1,"maker":{"key":"liquid","name":"Liquid AI","logo_url":"https://handyai.news/ui/makers/liquid.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":34},{"model":"Nemotron 3.5 Lightning","score":52,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":52,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-lightning","rank":35},{"model":"gpt-oss-20b","score":32.88,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":36},{"model":"Nemotron 3 Nano","score":32.75,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":35.9,"reports":2,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":37}]},{"key":"programbench","name":"ProgramBench","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#programbench","api_url":"https://handyai.news/api/benchmarks/programbench","image_url":"https://handyai.news/api/benchmarks/programbench.png","leader":{"model":"Claude Opus 5.5","score":65,"maker":"Anthropic"},"model_count":14,"drop_count":5,"first_date":"2026-06-26","last_date":"2026-11-04","frontier_gain":42,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":23,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":37,"maker":"anthropic"},{"date":"2026-09-01","model":"Fable 5.1","score":53.5,"maker":"anthropic"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":65,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":65,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"Muse Wonder Full","score":59,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":2},{"model":"Fable 5.1","score":53.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":3},{"model":"Muse Wonder","score":50,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":50,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":4},{"model":"GPT-6 Astra","score":50,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":5},{"model":"Claude Opus 5","score":37,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":6},{"model":"Kimi K3.1 Pro","score":31.5,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":31.5,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":7},{"model":"Hy4","score":30,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Tencent Hunyuan Hy4 launch post","slug":"hy4","url":"https://x.com/TencentHunyuan/status/2117870250838630539","published":"2026-11-04"},"launch_score":30,"reports":1,"maker":{"key":"tencent","name":"Tencent","logo_url":"https://handyai.news/ui/makers/tencent.svg"},"open_weights":true,"released":"2026-11-04","dated_by":"own_drop","drop_url":null,"rank":8},{"model":"DeepSeek V4.1 Flash 1017","score":29.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":29.1,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":9},{"model":"LongCat-2.5","score":25.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Meituan LongCat-2.5 launch post","slug":"longcat-2-5","url":"https://x.com/Meituan_LongCat/status/2117264311714537633","published":"2026-11-02"},"launch_score":25.5,"reports":1,"maker":{"key":"meituan","name":"Meituan","logo_url":"https://handyai.news/ui/makers/meituan.svg"},"open_weights":true,"released":"2026-11-02","dated_by":"own_drop","drop_url":null,"rank":10},{"model":"GPT-5.6 Sol","score":23,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":11},{"model":"DeepSeek V4.1 Flash","score":20.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":12},{"model":"GLM-5.3","score":19,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":13},{"model":"Kimi K3","score":17.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":14}]},{"key":"cybergym","name":"CyberGym","about":null,"type":{"key":"cyber","label":"Cyber"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#cybergym","api_url":"https://handyai.news/api/benchmarks/cybergym","image_url":"https://handyai.news/api/benchmarks/cybergym.png","leader":{"model":"MAI-Cyber-1-Flash","score":95.95,"maker":"Microsoft"},"model_count":13,"drop_count":5,"first_date":"2025-12-01","last_date":"2026-10-19","frontier_gain":78.7,"frontier":[{"date":"2025-12-01","model":"DeepSeek V3.2","score":17.3,"maker":"deepseek"},{"date":"2026-01-27","model":"Kimi K2.5","score":41.3,"maker":"moonshot"},{"date":"2026-02-05","model":"Claude Opus 4.6","score":66.6,"maker":"anthropic"},{"date":"2026-03-27","model":"GLM-5.1","score":68.7,"maker":"zai"},{"date":"2026-04-07","model":"Claude Mythos Preview","score":83.1,"maker":"anthropic"},{"date":"2026-06-09","model":"Mythos 5","score":83.8,"maker":"anthropic"},{"date":"2026-07-27","model":"MAI-Cyber-1-Flash","score":95.95,"maker":"microsoft"}],"models":[{"model":"MAI-Cyber-1-Flash","score":95.95,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Cyber-1-Flash","slug":"mai-cyber-1-flash","url":"https://handyai.news/modeldrop/mai-cyber-1-flash","published":"2026-07-27"},"launch_score":95.95,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-07-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-cyber-1-flash","rank":1},{"model":"GPT-5.5 Cyber","score":85.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Cyber-1-Flash","slug":"mai-cyber-1-flash","url":"https://handyai.news/modeldrop/mai-cyber-1-flash","published":"2026-07-27"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-07-27","dated_by":"first_cited","drop_url":null,"rank":2},{"model":"GLM-5.3","score":84.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":84.5,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":3},{"model":"Mythos 5","score":83.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":4},{"model":"GPT-5.6 Sol","score":83.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":5},{"model":"Gemini 3.5 Flash Cyber","score":83.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Cyber-1-Flash","slug":"mai-cyber-1-flash","url":"https://handyai.news/modeldrop/mai-cyber-1-flash","published":"2026-07-27"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-27","dated_by":"first_cited","drop_url":null,"rank":6},{"model":"Claude Mythos Preview","score":83.1,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Claude Mythos Preview","slug":"claude-mythos-preview","url":"https://handyai.news/modeldrop/claude-mythos-preview","published":"2026-04-07"},"launch_score":83.1,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":7},{"model":"MiniMax M3.1 Flash","score":82.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":82.2,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":8},{"model":"Kimi K3","score":80,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":9},{"model":"GLM-5.1","score":68.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":68.7,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-03-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-51","rank":10},{"model":"Claude Opus 4.6","score":66.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Mythos Preview","slug":"claude-mythos-preview","url":"https://handyai.news/modeldrop/claude-mythos-preview","published":"2026-04-07"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":11},{"model":"Kimi K2.5","score":41.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":12},{"model":"DeepSeek V3.2","score":17.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-12-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v32","rank":13}]},{"key":"math-500","name":"MATH-500","about":null,"type":{"key":"math","label":"Math"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#math-500","api_url":"https://handyai.news/api/benchmarks/math-500","image_url":"https://handyai.news/api/benchmarks/math-500.png","leader":{"model":"Nemotron Nano 2","score":97.8,"maker":"NVIDIA"},"model_count":13,"drop_count":5,"first_date":"2024-12-05","last_date":"2025-08-18","frontier_gain":1.4,"frontier":[{"date":"2024-12-05","model":"o1","score":96.4,"maker":"openai"},{"date":"2025-01-20","model":"DeepSeek R1","score":97.3,"maker":"deepseek"},{"date":"2025-08-18","model":"Nemotron Nano 2","score":97.8,"maker":"nvidia"}],"models":[{"model":"Nemotron Nano 2","score":97.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":97.8,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-nano-2","rank":1},{"model":"DeepSeek R1","score":97.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Llama 3.1 Nemotron Ultra","slug":"llama-31-nemotron-ultra","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","published":"2025-04-07"},"launch_score":97.3,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":2},{"model":"Llama 3.1 Nemotron Ultra","score":97,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Llama 3.1 Nemotron Ultra","slug":"llama-31-nemotron-ultra","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","published":"2025-04-07"},"launch_score":97,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","rank":3},{"model":"o1","score":96.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1","slug":"deepseek-r1","url":"https://handyai.news/modeldrop/deepseek-r1","published":"2025-01-20"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":4},{"model":"Qwen3-8B","score":96.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"first_cited","drop_url":null,"rank":5},{"model":"Mistral Medium 3","score":91,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":91,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2025-05-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-medium-3","rank":6},{"model":"DeepSeek V3","score":90.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":90.2,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v3","rank":7},{"model":"o1-mini","score":90,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1","slug":"deepseek-r1","url":"https://handyai.news/modeldrop/deepseek-r1","published":"2025-01-20"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-20","dated_by":"first_cited","drop_url":null,"rank":8},{"model":"Llama 4 Maverick","score":90,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":9},{"model":"Claude 3.7 Sonnet","score":83,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":10},{"model":"Claude 3.5 Sonnet","score":78.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V3","slug":"deepseek-v3","url":"https://handyai.news/modeldrop/deepseek-v3","published":"2024-12-26"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":11},{"model":"GPT-4o","score":76.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":12},{"model":"Llama 3.1 405B","score":66.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Llama 3.1 Nemotron Ultra","slug":"llama-31-nemotron-ultra","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","published":"2025-04-07"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"first_cited","drop_url":null,"rank":13}]},{"key":"swe-bench-multilingual","name":"SWE-bench Multilingual","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#swe-bench-multilingual","api_url":"https://handyai.news/api/benchmarks/swe-bench-multilingual","image_url":"https://handyai.news/api/benchmarks/swe-bench-multilingual.png","leader":{"model":"GPT-5.4","score":76.8,"maker":"OpenAI"},"model_count":13,"drop_count":5,"first_date":"2025-09-24","last_date":"2026-10-27","frontier_gain":20.9,"frontier":[{"date":"2025-09-24","model":"Kimi K2 0905","score":55.9,"maker":"moonshot"},{"date":"2025-09-29","model":"Claude Sonnet 4.5","score":68,"maker":"anthropic"},{"date":"2026-02-05","model":"Claude Opus 4.6","score":75.8,"maker":"anthropic"},{"date":"2026-03-05","model":"GPT-5.4","score":76.8,"maker":"openai"}],"models":[{"model":"GPT-5.4","score":76.8,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Composer 2","slug":"composer-2","url":"https://handyai.news/modeldrop/composer-2","published":"2026-03-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":1},{"model":"Claude Opus 4.6","score":75.8,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Composer 2","slug":"composer-2","url":"https://handyai.news/modeldrop/composer-2","published":"2026-03-19"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":2},{"model":"Composer 2","score":73.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Composer 2","slug":"composer-2","url":"https://handyai.news/modeldrop/composer-2","published":"2026-03-19"},"launch_score":73.7,"reports":1,"maker":{"key":"cursor","name":"Cursor","logo_url":"https://handyai.news/ui/makers/cursor.svg"},"open_weights":false,"released":"2026-03-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/composer-2","rank":3},{"model":"Claude Sonnet 4.5","score":68,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":4},{"model":"MAI-Code-1-Flash","score":65.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":65.5,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-06-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-code-1-flash","rank":5},{"model":"Kimi K2.5","score":65.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Composer 2","slug":"composer-2","url":"https://handyai.news/modeldrop/composer-2","published":"2026-03-19"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":6},{"model":"Claude Haiku 4.5","score":62.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":7},{"model":"DeepSeek V3.2-Exp","score":57.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":8},{"model":"Composer","score":56.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Composer","slug":"composer","url":"https://handyai.news/modeldrop/composer","published":"2025-10-29"},"launch_score":56.9,"reports":1,"maker":{"key":"cursor","name":"Cursor","logo_url":"https://handyai.news/ui/makers/cursor.svg"},"open_weights":false,"released":"2025-10-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/composer","rank":9},{"model":"MiniMax-M2","score":56.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":56.5,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m2","rank":10},{"model":"Kimi K2 0905","score":55.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-09-24","dated_by":"first_cited","drop_url":null,"rank":11},{"model":"GLM-4.6","score":53.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":12},{"model":"LFM3-1.7B","score":23,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Liquid AI, \"Introducing LFM3\"","slug":"lfm3-17b","url":"https://www.liquid.ai/blog/lfm3","published":"2026-10-27"},"launch_score":23,"reports":1,"maker":{"key":"liquid","name":"Liquid AI","logo_url":"https://handyai.news/ui/makers/liquid.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":13}]},{"key":"codeforces","name":"Codeforces","about":null,"type":{"key":"coding","label":"Coding"},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#codeforces","api_url":"https://handyai.news/api/benchmarks/codeforces","image_url":"https://handyai.news/api/benchmarks/codeforces.png","leader":{"model":"o3-pro","score":2748,"maker":"OpenAI"},"model_count":12,"drop_count":5,"first_date":"2024-12-05","last_date":"2025-08-05","frontier_gain":857,"frontier":[{"date":"2024-12-05","model":"o1","score":1891,"maker":"openai"},{"date":"2025-01-20","model":"DeepSeek R1","score":2029,"maker":"deepseek"},{"date":"2025-01-31","model":"o3-mini","score":2036,"maker":"openai"},{"date":"2025-04-16","model":"o4-mini","score":2719,"maker":"openai"},{"date":"2025-06-10","model":"o3-pro","score":2748,"maker":"openai"}],"models":[{"model":"o3-pro","score":2748,"own_drop":true,"independent":false,"setting":"","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":2748,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-06-10","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-pro","rank":1},{"model":"o4-mini","score":2719,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":2},{"model":"o3","score":2706,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":2706,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":3},{"model":"gpt-oss-120b","score":2622,"own_drop":true,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":2622,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":4},{"model":"gpt-oss-20b","score":2516,"own_drop":false,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":5},{"model":"Qwen3","score":2056,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":2056,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-04-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3","rank":6},{"model":"o3-mini","score":2036,"own_drop":false,"independent":false,"setting":"medium","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":7},{"model":"DeepSeek R1","score":2029,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":2029,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":8},{"model":"Gemini 2.5 Pro","score":2001,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":9},{"model":"o1","score":1891,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":10},{"model":"o1-mini","score":1820,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek R1","slug":"deepseek-r1","url":"https://handyai.news/modeldrop/deepseek-r1","published":"2025-01-20"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-20","dated_by":"first_cited","drop_url":null,"rank":11},{"model":"o1-pro","score":1707,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"first_cited","drop_url":null,"rank":12}]},{"key":"mmmu-pro","name":"MMMU-Pro","about":null,"type":{"key":"multimodal","label":"Multimodal"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#mmmu-pro","api_url":"https://handyai.news/api/benchmarks/mmmu-pro","image_url":"https://handyai.news/api/benchmarks/mmmu-pro.png","leader":{"model":"Claude Opus 5.5","score":87.7,"maker":"Anthropic"},"model_count":16,"drop_count":4,"first_date":"2025-09-29","last_date":"2026-10-23","frontier_gain":19.7,"frontier":[{"date":"2025-09-29","model":"Claude Sonnet 4.5","score":68,"maker":"anthropic"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":81,"maker":"google"},{"date":"2025-12-17","model":"Gemini 3 Flash","score":81.2,"maker":"google"},{"date":"2026-09-01","model":"Fable 5.1","score":86.3,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":86.9,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":87.7,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":87.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"GPT-6 Astra","score":86.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":2},{"model":"Fable 5.1","score":86.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":3},{"model":"Kimi K3.1 Pro","score":84.1,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":84.1,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":4},{"model":"Gemini 3 Flash","score":81.2,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":81.2,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-flash","rank":5},{"model":"Gemini 3 Pro","score":81,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":6},{"model":"DeepSeek V4.1 Flash 1017","score":80.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":7},{"model":"GPT-5.2","score":79.5,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":8},{"model":"Kimi K2.5","score":78.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":78.5,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":9},{"model":"Gemma 4","score":76.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":76.9,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":10},{"model":"Claude Opus 4.5","score":74,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":11},{"model":"Gemma 4 26B A4B","score":73.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":12},{"model":"Qwen3-VL-235B-A22B-Thinking","score":69.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"first_cited","drop_url":null,"rank":13},{"model":"Claude Sonnet 4.5","score":68,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":14},{"model":"Grok 4.1 Fast","score":63,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"first_cited","drop_url":null,"rank":15},{"model":"Gemma 3 27B","score":49.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"first_cited","drop_url":null,"rank":16}]},{"key":"nl2repo","name":"NL2Repo","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#nl2repo","api_url":"https://handyai.news/api/benchmarks/nl2repo","image_url":"https://handyai.news/api/benchmarks/nl2repo.png","leader":{"model":"Claude Opus 5","score":75.3,"maker":"Anthropic"},"model_count":15,"drop_count":4,"first_date":"2026-01-27","last_date":"2026-10-24","frontier_gain":43.3,"frontier":[{"date":"2026-01-27","model":"Kimi K2.5","score":32,"maker":"moonshot"},{"date":"2026-02-05","model":"Claude Opus 4.6","score":49.8,"maker":"anthropic"},{"date":"2026-06-26","model":"GPT-5.6 Sol","score":56.8,"maker":"openai"},{"date":"2026-07-16","model":"Kimi K3","score":58,"maker":"moonshot"},{"date":"2026-07-24","model":"Claude Opus 5","score":75.3,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5","score":75.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":1},{"model":"DeepSeek V4.1 Flash 1017","score":68.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":68.2,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":2},{"model":"DeepSeek V4.1 Flash","score":65.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":3},{"model":"MiniMax M3.1 Flash","score":61.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":61.2,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":4},{"model":"Kimi K3","score":58,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":5},{"model":"GLM-5.3","score":58,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":6},{"model":"GPT-5.6 Sol","score":56.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":7},{"model":"Nanbeige4.3-33B","score":53,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":53,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":8},{"model":"Claude Opus 4.6","score":49.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":9},{"model":"Qwen3.8","score":48.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":10},{"model":"GLM-5.1","score":42.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":42.7,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-03-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-51","rank":11},{"model":"GPT-5.4","score":41.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":12},{"model":"Gemini 3.1 Pro","score":33.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":13},{"model":"Kimi K2.5","score":32,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.1","slug":"glm-51","url":"https://handyai.news/modeldrop/glm-51","published":"2026-03-27"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":14},{"model":"Muse Glimmer-30B","score":27.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":15}]},{"key":"charxiv-reasoning","name":"CharXiv Reasoning","about":null,"type":{"key":"multimodal","label":"Multimodal"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#charxiv-reasoning","api_url":"https://handyai.news/api/benchmarks/charxiv-reasoning","image_url":"https://handyai.news/api/benchmarks/charxiv-reasoning.png","leader":{"model":"Muse Spark","score":86.4,"maker":"Meta"},"model_count":14,"drop_count":4,"first_date":"2026-02-05","last_date":"2026-10-15","frontier_gain":21.1,"frontier":[{"date":"2026-02-05","model":"Claude Opus 4.6","score":65.3,"maker":"anthropic"},{"date":"2026-02-19","model":"Gemini 3.1 Pro","score":80.2,"maker":"google"},{"date":"2026-03-05","model":"GPT-5.4","score":82.8,"maker":"openai"},{"date":"2026-03-11","model":"Qwen3.5-122B-A10B","score":83.47,"maker":"qwen"},{"date":"2026-04-08","model":"Muse Spark","score":86.4,"maker":"meta"}],"models":[{"model":"Muse Spark","score":86.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":86.4,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-04-08","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark","rank":1},{"model":"GPT-5.6 Terra","score":85.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":2},{"model":"Gemini 3.6 Flash","score":85.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":85.2,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":3},{"model":"Gemini 3.7 Flash","score":84.5,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":84.5,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-08-13","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-37-flash","rank":4},{"model":"Qwen3.5-122B-A10B","score":83.47,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"first_cited","drop_url":null,"rank":5},{"model":"GPT-5.4","score":82.8,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":6},{"model":"GPT-5.6 Luna","score":82.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":7},{"model":"GLM-5.3-Flash","score":81.93,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"first_cited","drop_url":null,"rank":8},{"model":"Grok 4.5","score":81.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":9},{"model":"Nemotron 3.5 Super VL","score":80.82,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":80.82,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":10},{"model":"Gemini 3.1 Pro","score":80.2,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":11},{"model":"Inkling Small","score":77.77,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"first_cited","drop_url":null,"rank":12},{"model":"Claude Sonnet 5","score":77,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":13},{"model":"Claude Opus 4.6","score":65.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":14}]},{"key":"osworld-2-offline-subset","name":"OSWorld 2 (offline subset)","about":"Using a computer, on the 82 OSWorld 2.1 tasks that need no internet.","type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#osworld-2-offline-subset","api_url":"https://handyai.news/api/benchmarks/osworld-2-offline-subset","image_url":"https://handyai.news/api/benchmarks/osworld-2-offline-subset.png","leader":{"model":"Claude Opus 5.5","score":87.2,"maker":"Anthropic"},"model_count":13,"drop_count":4,"first_date":"2025-10-15","last_date":"2026-10-28","frontier_gain":71.5,"frontier":[{"date":"2025-10-15","model":"Claude Haiku 4.5","score":15.7,"maker":"anthropic"},{"date":"2026-06-26","model":"GPT-5.6 Sol","score":65.7,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":70.2,"maker":"anthropic"},{"date":"2026-09-01","model":"Fable 5.1","score":76.8,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":80.9,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":87.2,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":87.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"Claude Sonnet 5.5","score":83.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":2},{"model":"Gemini 4 Carbon","score":81.5,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":81.5,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":3},{"model":"GPT-6 Astra","score":80.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":4},{"model":"Fable 5.1","score":76.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":5},{"model":"Gemini 4 Argon","score":74.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":6},{"model":"Claude Haiku 5.5","score":72.4,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":72.4,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":7},{"model":"GPT-6.1 Sol","score":71.4,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Sol","slug":"gpt-61-sol","url":"https://handyai.news/modeldrop/gpt-61-sol","published":"2026-09-29"},"launch_score":71.4,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":8},{"model":"Claude Opus 5","score":70.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6 Sol & GPT-6 Luna","slug":"gpt-6-sol-and-gpt-6-luna","url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","published":"2026-09-22"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":9},{"model":"GPT-5.6 Sol","score":65.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-6.1 Sol","slug":"gpt-61-sol","url":"https://handyai.news/modeldrop/gpt-61-sol","published":"2026-09-29"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":10},{"model":"GPT-6 Sol","score":64.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Sol","slug":"gpt-61-sol","url":"https://handyai.news/modeldrop/gpt-61-sol","published":"2026-09-29"},"launch_score":64.4,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":11},{"model":"GPT-6 Luna","score":48.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":12},{"model":"Claude Haiku 4.5","score":15.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":13}]},{"key":"aa-omniscience-hallucination-rate","name":"AA-Omniscience hallucination rate","about":null,"type":{"key":"knowledge","label":"Knowledge"},"unit":"%","lower_is_better":true,"url":"https://handyai.news/benchmarks#aa-omniscience-hallucination-rate","api_url":"https://handyai.news/api/benchmarks/aa-omniscience-hallucination-rate","image_url":"https://handyai.news/api/benchmarks/aa-omniscience-hallucination-rate.png","leader":{"model":"Muse Wonder Full","score":23,"maker":"Meta"},"model_count":12,"drop_count":4,"first_date":"2025-11-06","last_date":"2026-10-26","frontier_gain":51,"frontier":[{"date":"2025-11-06","model":"Kimi K2 Thinking","score":74,"maker":"moonshot"},{"date":"2026-01-27","model":"Kimi K2.5","score":64,"maker":"moonshot"},{"date":"2026-08-12","model":"Grok 4.6","score":34,"maker":"xai"},{"date":"2026-09-21","model":"Grok 4.7","score":29,"maker":"xai"},{"date":"2026-10-12","model":"Muse Wonder Full","score":23,"maker":"meta"}],"models":[{"model":"Muse Wonder Full","score":23,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":1},{"model":"Muse Wonder","score":26.5,"own_drop":true,"independent":true,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":26.5,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":2},{"model":"Grok 4.7","score":29,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":29,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":3},{"model":"Grok 4.6","score":34,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":4},{"model":"GPT-6.1 Terra","score":38,"own_drop":true,"independent":true,"setting":"","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":38,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":5},{"model":"Claude Sonnet 5.5","score":47,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":6},{"model":"GPT-6 Astra","score":51.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":7},{"model":"Claude Opus 5.5","score":58.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":8},{"model":"Kimi K2.5","score":64,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":64,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":9},{"model":"Fable 5.1","score":72.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":10},{"model":"Kimi K2 Thinking","score":74,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-11-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2-thinking","rank":11},{"model":"GPT-5.6 Terra","score":88,"own_drop":false,"independent":true,"setting":"","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":12}]},{"key":"aime-2026","name":"AIME 2026","about":null,"type":{"key":"math","label":"Math"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#aime-2026","api_url":"https://handyai.news/api/benchmarks/aime-2026","image_url":"https://handyai.news/api/benchmarks/aime-2026.png","leader":{"model":"GLM-5.2","score":99.2,"maker":"Z.ai"},"model_count":12,"drop_count":4,"first_date":"2025-10-15","last_date":"2026-10-24","frontier_gain":15.9,"frontier":[{"date":"2025-10-15","model":"Claude Haiku 4.5","score":83.3,"maker":"anthropic"},{"date":"2026-04-02","model":"Gemma 4","score":89.2,"maker":"google"},{"date":"2026-04-22","model":"Kimi K2.6","score":96.4,"maker":"moonshot"},{"date":"2026-07-15","model":"GLM-5.2","score":99.2,"maker":"zai"}],"models":[{"model":"GLM-5.2","score":99.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"first_cited","drop_url":null,"rank":1},{"model":"Inkling","score":97.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":97.1,"reports":1,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":2},{"model":"Kimi K2.6","score":96.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-04-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k26","rank":3},{"model":"Nanbeige4.3-33B","score":95.2,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":95.2,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":4},{"model":"Muse Glimmer-30B","score":94.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":5},{"model":"Qwen3.8","score":94.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":6},{"model":"Nemotron 3 Ultra","score":94.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Inkling","slug":"inkling","url":"https://handyai.news/modeldrop/inkling","published":"2026-07-15"},"launch_score":null,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":7},{"model":"MAI-Code-1-Flash","score":92.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":92.5,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-06-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-code-1-flash","rank":8},{"model":"Gemma 4","score":89.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":89.2,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":9},{"model":"Gemma 4 26B A4B","score":88.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":10},{"model":"Claude Haiku 4.5","score":83.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":11},{"model":"Gemma 3 27B","score":20.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"first_cited","drop_url":null,"rank":12}]},{"key":"bench-retail","name":"τ²-bench Retail","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#bench-retail","api_url":"https://handyai.news/api/benchmarks/bench-retail","image_url":"https://handyai.news/api/benchmarks/bench-retail.png","leader":{"model":"Claude Opus 4.1","score":86.8,"maker":"Anthropic"},"model_count":12,"drop_count":4,"first_date":"2025-04-14","last_date":"2026-04-02","frontier_gain":12,"frontier":[{"date":"2025-04-14","model":"GPT-4.1","score":74.8,"maker":"openai"},{"date":"2025-05-22","model":"Claude Sonnet 4","score":83.8,"maker":"anthropic"},{"date":"2025-08-05","model":"Claude Opus 4.1","score":86.8,"maker":"anthropic"}],"models":[{"model":"Claude Opus 4.1","score":86.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":1},{"model":"Gemma 4","score":86.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":86.4,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":2},{"model":"Claude Sonnet 4.5","score":86.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":86.2,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":3},{"model":"Gemma 4 26B A4B","score":85.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":4},{"model":"Claude Sonnet 4","score":83.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":5},{"model":"Claude Opus 4","score":81.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":6},{"model":"GPT-5","score":81.1,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.1","slug":"gpt-51","url":"https://handyai.news/modeldrop/gpt-51","published":"2025-11-12"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":7},{"model":"GPT-5.1","score":77.9,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"GPT-5.1","slug":"gpt-51","url":"https://handyai.news/modeldrop/gpt-51","published":"2025-11-12"},"launch_score":77.9,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":8},{"model":"GPT-4.1","score":74.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":9},{"model":"Kimi K2","score":70.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":70.6,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-07-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2","rank":10},{"model":"DeepSeek V3 0324","score":69.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2","slug":"kimi-k2","url":"https://handyai.news/modeldrop/kimi-k2","published":"2025-07-11"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-04-14","dated_by":"first_cited","drop_url":null,"rank":11},{"model":"Gemma 3 27B","score":6.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"first_cited","drop_url":null,"rank":12}]},{"key":"agents-last-exam","name":"Agents' Last Exam","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#agents-last-exam","api_url":"https://handyai.news/api/benchmarks/agents-last-exam","image_url":"https://handyai.news/api/benchmarks/agents-last-exam.png","leader":{"model":"Kimi K3.1 Pro","score":43.2,"maker":"Moonshot AI"},"model_count":11,"drop_count":4,"first_date":"2026-06-26","last_date":"2026-11-04","frontier_gain":12.6,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":30.6,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":32.2,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":32.8,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":42.6,"maker":"anthropic"},{"date":"2026-10-23","model":"Kimi K3.1 Pro","score":43.2,"maker":"moonshot"}],"models":[{"model":"Kimi K3.1 Pro","score":43.2,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":43.2,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":1},{"model":"Claude Opus 5.5","score":42.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2},{"model":"Hy4","score":41,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Tencent Hunyuan Hy4 launch post","slug":"hy4","url":"https://x.com/TencentHunyuan/status/2117870250838630539","published":"2026-11-04"},"launch_score":41,"reports":1,"maker":{"key":"tencent","name":"Tencent","logo_url":"https://handyai.news/ui/makers/tencent.svg"},"open_weights":true,"released":"2026-11-04","dated_by":"own_drop","drop_url":null,"rank":3},{"model":"DeepSeek V4.1 Flash 1017","score":35.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":35.4,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":4},{"model":"GPT-6 Astra","score":32.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":5},{"model":"Claude Opus 5","score":32.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":6},{"model":"DeepSeek V4.1 Flash","score":31.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":7},{"model":"GPT-5.6 Sol","score":30.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":8},{"model":"MiniMax M3.1 Flash","score":29.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":29.6,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":9},{"model":"Kimi K3","score":28.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":10},{"model":"GLM-5.3","score":26.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":11}]},{"key":"jobbench","name":"JobBench","about":null,"type":{"key":"work","label":"Work"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#jobbench","api_url":"https://handyai.news/api/benchmarks/jobbench","image_url":"https://handyai.news/api/benchmarks/jobbench.png","leader":{"model":"Claude Opus 5","score":65.7,"maker":"Anthropic"},"model_count":11,"drop_count":4,"first_date":"2026-04-02","last_date":"2026-10-24","frontier_gain":56.7,"frontier":[{"date":"2026-04-02","model":"Gemma 4","score":9,"maker":"google"},{"date":"2026-06-26","model":"GPT-5.6 Sol","score":45.4,"maker":"openai"},{"date":"2026-07-16","model":"Kimi K3","score":54.3,"maker":"moonshot"},{"date":"2026-07-24","model":"Claude Opus 5","score":65.7,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5","score":65.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":1},{"model":"Muse Spark 1.3","score":64.9,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":64.9,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-13","rank":2},{"model":"MiMo-V2.6-Pro","score":62,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":62,"reports":1,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":3},{"model":"Muse Spark Open","score":61.8,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":61.8,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":4},{"model":"GLM-5.3","score":61.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":5},{"model":"Kimi K3","score":54.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":6},{"model":"Qwen3.8-2.4T-A95B","score":52.7,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":7},{"model":"GPT-5.6 Sol","score":45.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":8},{"model":"Nanbeige4.3-33B","score":44.8,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":44.8,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":9},{"model":"Qwen3.8","score":34.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":10},{"model":"Gemma 4","score":9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":11}]},{"key":"ifeval","name":"IFEval","about":null,"type":{"key":"instructions","label":"Instructions"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#ifeval","api_url":"https://handyai.news/api/benchmarks/ifeval","image_url":"https://handyai.news/api/benchmarks/ifeval.png","leader":{"model":"LFM3-1.7B","score":96.66,"maker":"Liquid AI"},"model_count":9,"drop_count":4,"first_date":"2024-12-05","last_date":"2026-10-27","frontier_gain":6.5,"frontier":[{"date":"2024-12-05","model":"Claude 3.5 Sonnet","score":90.2,"maker":"anthropic"},{"date":"2025-02-24","model":"Claude 3.7 Sonnet","score":93.2,"maker":"anthropic"},{"date":"2026-10-27","model":"LFM3-1.7B","score":96.66,"maker":"liquid"}],"models":[{"model":"LFM3-1.7B","score":96.66,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Liquid AI, \"Introducing LFM3\"","slug":"lfm3-17b","url":"https://www.liquid.ai/blog/lfm3","published":"2026-10-27"},"launch_score":96.66,"reports":1,"maker":{"key":"liquid","name":"Liquid AI","logo_url":"https://handyai.news/ui/makers/liquid.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":1},{"model":"Claude 3.7 Sonnet","score":93.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude 3.7 Sonnet","slug":"claude-37-sonnet","url":"https://handyai.news/modeldrop/claude-37-sonnet","published":"2025-02-24"},"launch_score":93.2,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":2},{"model":"Nemotron Nano 2","score":90.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":90.3,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-nano-2","rank":3},{"model":"Claude 3.5 Sonnet","score":90.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude 3.7 Sonnet","slug":"claude-37-sonnet","url":"https://handyai.news/modeldrop/claude-37-sonnet","published":"2025-02-24"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":4},{"model":"Qwen3-8B","score":89.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"first_cited","drop_url":null,"rank":5},{"model":"GPT-4.5","score":88.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":6},{"model":"GPT-4.1","score":87.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":87.4,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":7},{"model":"DeepSeek R1","score":83.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude 3.7 Sonnet","slug":"claude-37-sonnet","url":"https://handyai.news/modeldrop/claude-37-sonnet","published":"2025-02-24"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":8},{"model":"GPT-4o","score":81,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":9}]},{"key":"mmlu","name":"MMLU","about":null,"type":{"key":"knowledge","label":"Knowledge"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#mmlu","api_url":"https://handyai.news/api/benchmarks/mmlu","image_url":"https://handyai.news/api/benchmarks/mmlu.png","leader":{"model":"o3","score":93.4,"maker":"OpenAI"},"model_count":9,"drop_count":4,"first_date":"2024-12-05","last_date":"2026-10-28","frontier_gain":1.6,"frontier":[{"date":"2024-12-05","model":"o1","score":91.8,"maker":"openai"},{"date":"2025-04-16","model":"o3","score":93.4,"maker":"openai"}],"models":[{"model":"o3","score":93.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":1},{"model":"o4-mini","score":93,"own_drop":false,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":2},{"model":"o1","score":91.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"o1","slug":"o1","url":"https://handyai.news/modeldrop/o1","published":"2024-12-05"},"launch_score":91.8,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":3},{"model":"o1-preview","score":90.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o1","slug":"o1","url":"https://handyai.news/modeldrop/o1","published":"2024-12-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":4},{"model":"gpt-oss-120b","score":90,"own_drop":true,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":90,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":5},{"model":"o3-mini","score":87,"own_drop":false,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":86.9,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":6},{"model":"Particle-1","score":86.08,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Particle, \"Introducing Particle-1\"","slug":"particle-1","url":"https://particle.ai/blog/introducing-particle-1","published":"2026-10-28"},"launch_score":86.08,"reports":1,"maker":{"key":"particle","name":"Particle","logo_url":"https://handyai.news/ui/makers/particle.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":null,"rank":7},{"model":"gpt-oss-20b","score":85.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"gpt-oss-120b & gpt-oss-20b","slug":"gpt-oss","url":"https://handyai.news/modeldrop/gpt-oss","published":"2025-08-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":8},{"model":"o1-mini","score":85.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-20","dated_by":"first_cited","drop_url":null,"rank":9}]},{"key":"terminal-bench-4-0-standard-harness","name":"Terminal-Bench 4.0 (standard harness)","about":"Terminal-Bench run by a third party on one shared harness.","type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#terminal-bench-4-0-standard-harness","api_url":"https://handyai.news/api/benchmarks/terminal-bench-4-0-standard-harness","image_url":"https://handyai.news/api/benchmarks/terminal-bench-4-0-standard-harness.png","leader":{"model":"Kimi K3.1 Pro","score":67.2,"maker":"Moonshot AI"},"model_count":29,"drop_count":3,"first_date":"2026-07-15","last_date":"2026-10-28","frontier_gain":66.2,"frontier":[{"date":"2026-07-15","model":"Inkling","score":1,"maker":"thinkingmachines"},{"date":"2026-07-16","model":"Kimi K3","score":12.6,"maker":"moonshot"},{"date":"2026-07-19","model":"Qwen3.8-Max","score":38.9,"maker":"qwen"},{"date":"2026-08-14","model":"GLM-5.3","score":41.9,"maker":"zai"},{"date":"2026-09-01","model":"Fable 5.1","score":52,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":59.1,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":59.6,"maker":"anthropic"},{"date":"2026-09-28","model":"Claude Sonnet 5.5","score":63.6,"maker":"anthropic"},{"date":"2026-10-23","model":"Kimi K3.1 Pro","score":67.2,"maker":"moonshot"}],"models":[{"model":"Kimi K3.1 Pro","score":67.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":1},{"model":"Gemini 4 Carbon","score":65.2,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":65.2,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":2},{"model":"Claude Sonnet 5.5","score":63.6,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":3},{"model":"Claude Opus 5.5","score":59.6,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":4},{"model":"GPT-6 Astra","score":59.1,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":5},{"model":"Gemini 4 Argon","score":57.1,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":6},{"model":"Muse Wonder Full","score":57.1,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":7},{"model":"GPT-6.1 Sol","score":56.1,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":8},{"model":"GPT-6.1 Terra","score":55.6,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":9},{"model":"Fable 5.1","score":52,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":10},{"model":"Muse Wonder","score":52,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":11},{"model":"GLM-5.3","score":41.9,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":12},{"model":"Qwen3.8-Max","score":38.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":13},{"model":"MiMo-V2.6-Pro","score":34.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":14},{"model":"DeepSeek V4.1 Flash 1017","score":33.8,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":15},{"model":"Ling-3.1-flash-1019","score":32.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"antgroup","name":"Ant Group","logo_url":"https://handyai.news/ui/makers/antgroup.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":16},{"model":"Claude Haiku 5.5","score":32.8,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":17},{"model":"DeepSeek V4.1 Flash","score":27,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":18},{"model":"Grok 4.7","score":25.8,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":26,"reports":3,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":19},{"model":"MiniMax M3.1 Flash","score":25.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":20},{"model":"Mistral Large 4","score":24.7,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":22.73,"reports":3,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":21},{"model":"Gemini 3.8 Flash","score":19.7,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":22},{"model":"DeepSeek V4 Pro 0813","score":14.14,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"first_cited","drop_url":null,"rank":23},{"model":"Kimi K3","score":12.6,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":24},{"model":"GPT-6 Luna","score":12.6,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":25},{"model":"Muse Spark Open","score":8.1,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":26},{"model":"Qwen3.8","score":5.6,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":27},{"model":"Inkling","score":1,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":28},{"model":"Nemotron 3.5 Super VL","score":0,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":29}]},{"key":"gdp-pdf","name":"GDP.pdf","about":"Work products built from PDFs (Surge AI).","type":{"key":"work","label":"Work"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#gdp-pdf","api_url":"https://handyai.news/api/benchmarks/gdp-pdf","image_url":"https://handyai.news/api/benchmarks/gdp-pdf.png","leader":{"model":"Kimi K3.1 Pro","score":34.6,"maker":"Moonshot AI"},"model_count":13,"drop_count":3,"first_date":"2026-06-26","last_date":"2026-10-23","frontier_gain":9.9,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Terra","score":24.7,"maker":"openai"},{"date":"2026-06-30","model":"Claude Sonnet 5","score":28,"maker":"anthropic"},{"date":"2026-08-13","model":"Gemini 3.7 Flash","score":34,"maker":"google"},{"date":"2026-09-03","model":"GPT-6 Astra","score":34.2,"maker":"openai"},{"date":"2026-10-23","model":"Kimi K3.1 Pro","score":34.6,"maker":"moonshot"}],"models":[{"model":"Kimi K3.1 Pro","score":34.6,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":34.6,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":1},{"model":"GPT-6 Astra","score":34.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":2},{"model":"Gemini 3.7 Flash","score":34,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":34,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-08-13","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-37-flash","rank":3},{"model":"GPT-6.1 Sol","score":32,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"GPT-6.1 Sol","slug":"gpt-61-sol","url":"https://handyai.news/modeldrop/gpt-61-sol","published":"2026-09-29"},"launch_score":32,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":4},{"model":"Claude Opus 5.5","score":30.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":5},{"model":"Claude Sonnet 5","score":28,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":6},{"model":"GPT-6 Sol","score":28,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-6.1 Sol","slug":"gpt-61-sol","url":"https://handyai.news/modeldrop/gpt-61-sol","published":"2026-09-29"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":7},{"model":"Fable 5.1","score":27.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":8},{"model":"GPT-5.6 Terra","score":24.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":9},{"model":"Gemini 3.6 Flash","score":22,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":10},{"model":"DeepSeek V4.1 Flash 1017","score":17.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":11},{"model":"Muse Spark 1.2","score":16,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-12","rank":12},{"model":"GLM-5.3","score":11.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":13}]},{"key":"gpqa","name":"GPQA","about":null,"type":{"key":"reasoning","label":"Reasoning"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#gpqa","api_url":"https://handyai.news/api/benchmarks/gpqa","image_url":"https://handyai.news/api/benchmarks/gpqa.png","leader":{"model":"GPT-5.2","score":92.4,"maker":"OpenAI"},"model_count":13,"drop_count":3,"first_date":"2024-12-05","last_date":"2026-02-16","frontier_gain":38.8,"frontier":[{"date":"2024-12-05","model":"GPT-4o","score":53.6,"maker":"openai"},{"date":"2025-01-31","model":"o3-mini","score":79.7,"maker":"openai"},{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":84.4,"maker":"google"},{"date":"2025-07-09","model":"Grok 4","score":87.7,"maker":"xai"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":91.9,"maker":"google"},{"date":"2025-12-11","model":"GPT-5.2","score":92.4,"maker":"openai"}],"models":[{"model":"GPT-5.2","score":92.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":1},{"model":"Gemini 3 Pro","score":91.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":2},{"model":"Qwen3.5","score":88.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":88.4,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-02-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen35","rank":3},{"model":"Grok 4","score":87.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":4},{"model":"Kimi K2.5","score":87.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":5},{"model":"Claude Opus 4.5","score":87,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3.5","slug":"qwen35","url":"https://handyai.news/modeldrop/qwen35","published":"2026-02-16"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":6},{"model":"Gemini 2.5 Pro","score":84.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":7},{"model":"o3","score":82.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":8},{"model":"DeepSeek R1 0528","score":81.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1-0528","rank":9},{"model":"o3-mini","score":79.7,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":10},{"model":"GLM-4.5","score":79.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-4.5","slug":"glm-45","url":"https://handyai.news/modeldrop/glm-45","published":"2025-07-28"},"launch_score":79.1,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-07-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-45","rank":11},{"model":"GPT-4.5","score":71.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":71.4,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":12},{"model":"GPT-4o","score":53.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":13}]},{"key":"exploitbench","name":"ExploitBench","about":null,"type":{"key":"cyber","label":"Cyber"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#exploitbench","api_url":"https://handyai.news/api/benchmarks/exploitbench","image_url":"https://handyai.news/api/benchmarks/exploitbench.png","leader":{"model":"GPT-6 Astra","score":100,"maker":"OpenAI"},"model_count":11,"drop_count":3,"first_date":"2026-04-07","last_date":"2026-10-23","frontier_gain":31,"frontier":[{"date":"2026-04-07","model":"Claude Mythos Preview","score":69,"maker":"anthropic"},{"date":"2026-06-09","model":"Mythos 5","score":78,"maker":"anthropic"},{"date":"2026-09-01","model":"Fable 5.1","score":83,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":100,"maker":"openai"}],"models":[{"model":"GPT-6 Astra","score":100,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":1},{"model":"Claude Opus 5.5","score":91,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2},{"model":"Fable 5.1","score":83,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":3},{"model":"Mythos 5","score":78,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":78,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":4},{"model":"GPT-5.6 Sol","score":76.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":5},{"model":"Kimi K3.1 Pro","score":74.7,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":74.7,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":6},{"model":"Claude Mythos Preview","score":69,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":7},{"model":"GLM-5.3","score":54.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":54.4,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":8},{"model":"Claude Opus 4.8","score":40,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-48","rank":9},{"model":"GPT-5.5","score":34,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-04-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-55","rank":10},{"model":"Kimi K3","score":32.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":11}]},{"key":"terminal-bench-3-0","name":"Terminal-Bench 3.0","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#terminal-bench-3-0","api_url":"https://handyai.news/api/benchmarks/terminal-bench-3-0","image_url":"https://handyai.news/api/benchmarks/terminal-bench-3-0.png","leader":{"model":"Claude Opus 5","score":43.5,"maker":"Anthropic"},"model_count":11,"drop_count":3,"first_date":"2026-06-09","last_date":"2026-08-14","frontier_gain":9.8,"frontier":[{"date":"2026-06-09","model":"Fable 5","score":33.7,"maker":"anthropic"},{"date":"2026-06-26","model":"GPT-5.6 Sol","score":34.6,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":43.5,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5","score":43.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":1},{"model":"GPT-5.6 Sol","score":34.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":2},{"model":"Fable 5","score":33.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":3},{"model":"GLM-5.3","score":28.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":28.3,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":4},{"model":"Grok 4.6","score":26,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":26,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":5},{"model":"GPT-5.6 Terra","score":20.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":6},{"model":"Kimi K3","score":17.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":7},{"model":"Grok 4.5","score":15.7,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":8},{"model":"Gemini 3.7 Flash","score":14.9,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":14.9,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-08-13","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-37-flash","rank":9},{"model":"Claude Sonnet 5","score":14.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":10},{"model":"Gemini 3.6 Flash","score":5.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":11}]},{"key":"zerobench-with-tools","name":"ZeroBench (with tools)","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#zerobench-with-tools","api_url":"https://handyai.news/api/benchmarks/zerobench-with-tools","image_url":"https://handyai.news/api/benchmarks/zerobench-with-tools.png","leader":{"model":"Kimi K3.1 Pro","score":63,"maker":"Moonshot AI"},"model_count":11,"drop_count":3,"first_date":"2026-06-26","last_date":"2026-10-23","frontier_gain":10,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":53,"maker":"openai"},{"date":"2026-09-01","model":"Fable 5.1","score":59,"maker":"anthropic"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":61,"maker":"anthropic"},{"date":"2026-10-23","model":"Kimi K3.1 Pro","score":63,"maker":"moonshot"}],"models":[{"model":"Kimi K3.1 Pro","score":63,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":63,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":1},{"model":"Claude Opus 5.5","score":61,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2},{"model":"Fable 5.1","score":59,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":3},{"model":"GPT-6 Astra","score":58,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":4},{"model":"GPT-5.6 Sol","score":53,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":5},{"model":"Claude Opus 5","score":52,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":6},{"model":"DeepSeek V4.1 Flash 1017","score":51,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":51,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":7},{"model":"Muse Spark Open","score":50,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":50,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":8},{"model":"DeepSeek V4.1 Flash","score":49,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":9},{"model":"Kimi K3","score":41,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":10},{"model":"Inkling","score":14,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":11}]},{"key":"babyvision-with-tools","name":"BabyVision (with tools)","about":null,"type":{"key":"multimodal","label":"Multimodal"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#babyvision-with-tools","api_url":"https://handyai.news/api/benchmarks/babyvision-with-tools","image_url":"https://handyai.news/api/benchmarks/babyvision-with-tools.png","leader":{"model":"Claude Opus 5.5","score":97.7,"maker":"Anthropic"},"model_count":10,"drop_count":3,"first_date":"2026-06-26","last_date":"2026-10-28","frontier_gain":8.8,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":88.9,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":94.1,"maker":"anthropic"},{"date":"2026-09-01","model":"Fable 5.1","score":95.1,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":96.6,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":97.7,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":97.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"GPT-6 Astra","score":96.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":2},{"model":"Kimi K3.1 Pro","score":95.4,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":95.4,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":3},{"model":"Fable 5.1","score":95.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":4},{"model":"Claude Opus 5","score":94.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":5},{"model":"DeepSeek V4.1 Flash 1017","score":92.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":92.5,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":6},{"model":"DeepSeek V4.1 Flash","score":89.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":7},{"model":"Particle-1","score":89.18,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Particle, \"Introducing Particle-1\"","slug":"particle-1","url":"https://particle.ai/blog/introducing-particle-1","published":"2026-10-28"},"launch_score":89.18,"reports":1,"maker":{"key":"particle","name":"Particle","logo_url":"https://handyai.news/ui/makers/particle.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":null,"rank":8},{"model":"GPT-5.6 Sol","score":88.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":9},{"model":"Kimi K3","score":85.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":10}]},{"key":"frontiercode-v1-1","name":"FrontierCode v1.1","about":"Hard, long coding tasks.","type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#frontiercode-v1-1","api_url":"https://handyai.news/api/benchmarks/frontiercode-v1-1","image_url":"https://handyai.news/api/benchmarks/frontiercode-v1-1.png","leader":{"model":"Claude Opus 5.5","score":54.4,"maker":"Anthropic"},"model_count":9,"drop_count":3,"first_date":"2026-06-30","last_date":"2026-10-07","frontier_gain":12,"frontier":[{"date":"2026-06-30","model":"Claude Sonnet 5","score":42.4,"maker":"anthropic"},{"date":"2026-07-24","model":"Claude Opus 5","score":48,"maker":"anthropic"},{"date":"2026-09-01","model":"Fable 5.1","score":50.3,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":53.3,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":54.4,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":54.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 5.5","slug":"claude-sonnet-55","url":"https://handyai.news/modeldrop/claude-sonnet-55","published":"2026-09-28"},"launch_score":54.4,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"GPT-6 Astra","score":53.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 5.5","slug":"claude-opus-55","url":"https://handyai.news/modeldrop/claude-opus-55","published":"2026-09-22"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":2},{"model":"Claude Sonnet 5.5","score":52.1,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":46.2,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":3},{"model":"Fable 5.1","score":50.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 5.5","slug":"claude-opus-55","url":"https://handyai.news/modeldrop/claude-opus-55","published":"2026-09-22"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":4},{"model":"GPT-6 Sol","score":49.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 5.5","slug":"claude-sonnet-55","url":"https://handyai.news/modeldrop/claude-sonnet-55","published":"2026-09-28"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":5},{"model":"Claude Opus 5","score":48,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 5.5","slug":"claude-opus-55","url":"https://handyai.news/modeldrop/claude-opus-55","published":"2026-09-22"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":6},{"model":"Claude Haiku 5.5","score":46.4,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":46.4,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":7},{"model":"Claude Sonnet 5","score":42.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 5.5","slug":"claude-sonnet-55","url":"https://handyai.news/modeldrop/claude-sonnet-55","published":"2026-09-28"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":8},{"model":"GPT-6 Luna","score":42.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":9}]},{"key":"scicode","name":"SciCode","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#scicode","api_url":"https://handyai.news/api/benchmarks/scicode","image_url":"https://handyai.news/api/benchmarks/scicode.png","leader":{"model":"Ling-3.1-flash-1019","score":57,"maker":"Ant Group"},"model_count":6,"drop_count":3,"first_date":"2026-04-02","last_date":"2026-10-27","frontier_gain":11.5,"frontier":[{"date":"2026-04-02","model":"Gemma 4","score":45.5,"maker":"google"},{"date":"2026-08-12","model":"Qwen3.8","score":46.7,"maker":"qwen"},{"date":"2026-10-24","model":"Nanbeige4.3-33B","score":52.6,"maker":"nanbeige"},{"date":"2026-10-27","model":"Ling-3.1-flash-1019","score":57,"maker":"antgroup"}],"models":[{"model":"Ling-3.1-flash-1019","score":57,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Ling-3.1-flash-1019 model card","slug":"ling-31-flash-1019","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash-1019","published":"2026-10-27"},"launch_score":57,"reports":1,"maker":{"key":"antgroup","name":"Ant Group","logo_url":"https://handyai.news/ui/makers/antgroup.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":1},{"model":"Nanbeige4.3-33B","score":52.6,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":52.6,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":2},{"model":"Qwen3.8","score":46.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":3},{"model":"Gemma 4","score":45.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":4},{"model":"Muse Glimmer-30B","score":44.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":5},{"model":"LFM3-1.7B","score":28.47,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Liquid AI, \"Introducing LFM3\"","slug":"lfm3-17b","url":"https://www.liquid.ai/blog/lfm3","published":"2026-10-27"},"launch_score":28.47,"reports":1,"maker":{"key":"liquid","name":"Liquid AI","logo_url":"https://handyai.news/ui/makers/liquid.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":6}]},{"key":"tau3-bench","name":"Tau3-Bench","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#tau3-bench","api_url":"https://handyai.news/api/benchmarks/tau3-bench","image_url":"https://handyai.news/api/benchmarks/tau3-bench.png","leader":{"model":"LFM3-1.7B","score":68.01,"maker":"Liquid AI"},"model_count":6,"drop_count":3,"first_date":"2026-04-02","last_date":"2026-10-28","frontier_gain":49.4,"frontier":[{"date":"2026-04-02","model":"Gemma 4","score":18.6,"maker":"google"},{"date":"2026-08-12","model":"Qwen3.8","score":48.5,"maker":"qwen"},{"date":"2026-10-24","model":"Nanbeige4.3-33B","score":48.8,"maker":"nanbeige"},{"date":"2026-10-27","model":"LFM3-1.7B","score":68.01,"maker":"liquid"}],"models":[{"model":"LFM3-1.7B","score":68.01,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Liquid AI, \"Introducing LFM3\"","slug":"lfm3-17b","url":"https://www.liquid.ai/blog/lfm3","published":"2026-10-27"},"launch_score":68.01,"reports":1,"maker":{"key":"liquid","name":"Liquid AI","logo_url":"https://handyai.news/ui/makers/liquid.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":1},{"model":"Clover-35B-A3B-Preview","score":65.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DiffuSpace, \"Clover-35B-A3B-Preview\"","slug":"clover-35b-a3b-preview","url":"https://diffuspace.ai/blog/clover-35b/","published":"2026-10-28"},"launch_score":65.3,"reports":1,"maker":{"key":"diffuspace","name":"DiffuSpace","logo_url":"https://handyai.news/ui/makers/diffuspace.svg"},"open_weights":true,"released":"2026-10-28","dated_by":"own_drop","drop_url":null,"rank":2},{"model":"Nanbeige4.3-33B","score":48.8,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":48.8,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":3},{"model":"Qwen3.8","score":48.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":4},{"model":"Muse Glimmer-30B","score":33.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":5},{"model":"Gemma 4","score":18.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":6}]},{"key":"gdpval","name":"GDPval","about":"Work products judged against industry experts (win or tie rate).","type":{"key":"work","label":"Work"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#gdpval","api_url":"https://handyai.news/api/benchmarks/gdpval","image_url":"https://handyai.news/api/benchmarks/gdpval.png","leader":{"model":"GPT-5.4","score":83,"maker":"OpenAI"},"model_count":4,"drop_count":3,"first_date":"2025-08-07","last_date":"2026-03-05","frontier_gain":44.2,"frontier":[{"date":"2025-08-07","model":"GPT-5","score":38.8,"maker":"openai"},{"date":"2025-12-11","model":"GPT-5.2","score":70.9,"maker":"openai"},{"date":"2026-03-05","model":"GPT-5.4","score":83,"maker":"openai"}],"models":[{"model":"GPT-5.4","score":83,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":83,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":1},{"model":"GPT-5.2","score":70.9,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":70.9,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":2},{"model":"GPT-5.3-Codex","score":70.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.4","slug":"gpt-54","url":"https://handyai.news/modeldrop/gpt-54","published":"2026-03-05"},"launch_score":70.9,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-53-codex","rank":3},{"model":"GPT-5","score":38.8,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.2","slug":"gpt-52","url":"https://handyai.news/modeldrop/gpt-52","published":"2025-12-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":4}]},{"key":"aa-briefcase-v1-1","name":"AA-Briefcase v1.1","about":null,"type":{"key":"work","label":"Work"},"unit":"Elo","lower_is_better":false,"url":"https://handyai.news/benchmarks#aa-briefcase-v1-1","api_url":"https://handyai.news/api/benchmarks/aa-briefcase-v1-1","image_url":"https://handyai.news/api/benchmarks/aa-briefcase-v1-1.png","leader":{"model":"Claude Sonnet 5.5","score":1813,"maker":"Anthropic"},"model_count":29,"drop_count":2,"first_date":"2026-06-26","last_date":"2026-10-28","frontier_gain":326,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":1487,"maker":"openai"},{"date":"2026-07-16","model":"Kimi K3","score":1497,"maker":"moonshot"},{"date":"2026-07-19","model":"Qwen3.8-Max","score":1607,"maker":"qwen"},{"date":"2026-09-01","model":"Fable 5.1","score":1669,"maker":"anthropic"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":1800,"maker":"anthropic"},{"date":"2026-09-28","model":"Claude Sonnet 5.5","score":1813,"maker":"anthropic"}],"models":[{"model":"Claude Sonnet 5.5","score":1813,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":1},{"model":"Claude Opus 5.5","score":1800,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2},{"model":"Muse Wonder Full","score":1729,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":3},{"model":"Kimi K3.1 Pro","score":1675,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":1678,"reports":3,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":4},{"model":"Fable 5.1","score":1669,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":5},{"model":"Muse Wonder","score":1659,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":6},{"model":"Grok 4.7","score":1637,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":1657,"reports":3,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":7},{"model":"Gemini 4 Carbon","score":1610,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":8},{"model":"Qwen3.8-Max","score":1607,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":9},{"model":"Claude Haiku 5.5","score":1574,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":10},{"model":"DeepSeek V4.1 Flash 1017","score":1570,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":11},{"model":"GPT-6 Astra","score":1569,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":12},{"model":"GPT-6.1 Sol","score":1555,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":13},{"model":"Grok 4.6","score":1546,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":14},{"model":"Ling-3.1-flash-1019","score":1522,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"antgroup","name":"Ant Group","logo_url":"https://handyai.news/ui/makers/antgroup.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":15},{"model":"MiMo-V2.6-Pro","score":1512,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":16},{"model":"GLM-5.3","score":1503,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":17},{"model":"Kimi K3","score":1497,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":18},{"model":"GPT-5.6 Sol","score":1487,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":19},{"model":"Gemini 4 Argon","score":1484,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":20},{"model":"MiniMax M3.1 Flash","score":1475,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":21},{"model":"GPT-6.1 Terra","score":1401,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":22},{"model":"Qwen3.8","score":1393,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":23},{"model":"Mistral Large 4","score":1348,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":24},{"model":"GPT-6 Luna","score":1337,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":25},{"model":"Muse Spark Open","score":1278,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":26},{"model":"Gemini 3.8 Flash","score":1203,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":27},{"model":"Nemotron 3.5 Super VL","score":999,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":28},{"model":"Inkling","score":833,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":29}]},{"key":"automationbench-aa","name":"AutomationBench-AA","about":"Business automations, run by Artificial Analysis.","type":{"key":"work","label":"Work"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#automationbench-aa","api_url":"https://handyai.news/api/benchmarks/automationbench-aa","image_url":"https://handyai.news/api/benchmarks/automationbench-aa.png","leader":{"model":"Gemini 4 Carbon","score":77.8,"maker":"Google DeepMind"},"model_count":27,"drop_count":2,"first_date":"2026-07-15","last_date":"2026-10-28","frontier_gain":72.8,"frontier":[{"date":"2026-07-15","model":"Inkling","score":5,"maker":"thinkingmachines"},{"date":"2026-07-16","model":"Kimi K3","score":58.3,"maker":"moonshot"},{"date":"2026-08-14","model":"GLM-5.3","score":62.2,"maker":"zai"},{"date":"2026-09-03","model":"GPT-6 Astra","score":68.5,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":69.5,"maker":"anthropic"},{"date":"2026-09-28","model":"Claude Sonnet 5.5","score":71.8,"maker":"anthropic"},{"date":"2026-10-01","model":"Gemini 4 Argon","score":77.5,"maker":"google"},{"date":"2026-10-28","model":"Gemini 4 Carbon","score":77.8,"maker":"google"}],"models":[{"model":"Gemini 4 Carbon","score":77.8,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":1},{"model":"Gemini 4 Argon","score":77.5,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":2},{"model":"Kimi K3.1 Pro","score":73.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":73.3,"reports":3,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":3},{"model":"MiniMax M3.1 Flash","score":72.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":4},{"model":"Claude Sonnet 5.5","score":71.8,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":5},{"model":"Muse Wonder Full","score":71.3,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":6},{"model":"Claude Opus 5.5","score":69.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":7},{"model":"GPT-6 Astra","score":68.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":4,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":8},{"model":"Muse Wonder","score":68.1,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":39.1,"reports":3,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":9},{"model":"Ling-3.1-flash-1019","score":67.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"antgroup","name":"Ant Group","logo_url":"https://handyai.news/ui/makers/antgroup.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":10},{"model":"DeepSeek V4.1 Flash 1017","score":66.6,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":11},{"model":"Grok 4.7","score":65.6,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":12},{"model":"GPT-6.1 Sol","score":64.9,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":13},{"model":"GLM-5.3","score":62.2,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":14},{"model":"GPT-6.1 Terra","score":60.2,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":15},{"model":"Gemini 3.8 Flash","score":59.9,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":16},{"model":"Fable 5.1","score":59.4,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":4,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":17},{"model":"MiMo-V2.6-Pro","score":58.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":18},{"model":"Kimi K3","score":58.3,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":19},{"model":"Mistral Large 4","score":57.8,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":20},{"model":"Qwen3.8-Max","score":56.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":21},{"model":"GPT-6 Luna","score":53.2,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":22},{"model":"Muse Spark Open","score":49.7,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":23},{"model":"Qwen3.8","score":48.2,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":24},{"model":"Claude Haiku 5.5","score":35.4,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":25},{"model":"Nemotron 3.5 Super VL","score":14.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":26},{"model":"Inkling","score":5,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":27}]},{"key":"apex-agents","name":"APEX-Agents","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#apex-agents","api_url":"https://handyai.news/api/benchmarks/apex-agents","image_url":"https://handyai.news/api/benchmarks/apex-agents.png","leader":{"model":"Claude Opus 5.5","score":73.5,"maker":"Anthropic"},"model_count":11,"drop_count":2,"first_date":"2026-06-09","last_date":"2026-10-23","frontier_gain":14.3,"frontier":[{"date":"2026-06-09","model":"Fable 5","score":59.2,"maker":"anthropic"},{"date":"2026-07-24","model":"Claude Opus 5","score":60.6,"maker":"anthropic"},{"date":"2026-09-01","model":"Fable 5.1","score":68.6,"maker":"anthropic"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":73.5,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":73.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"Fable 5.1","score":68.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":2},{"model":"Kimi K3.1 Pro","score":67.2,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":67.2,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":3},{"model":"GPT-6 Astra","score":64.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":4},{"model":"DeepSeek V4.1 Flash 1017","score":63.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":5},{"model":"Claude Opus 5","score":60.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":6},{"model":"Fable 5","score":59.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":7},{"model":"Grok 4.6","score":57.5,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":57.5,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":8},{"model":"GPT-5.6 Sol","score":56.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":9},{"model":"GLM-5.3","score":56.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":10},{"model":"Grok 4.5","score":47.1,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":11}]},{"key":"agents-last-exam-v1","name":"Agents' Last Exam v1","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#agents-last-exam-v1","api_url":"https://handyai.news/api/benchmarks/agents-last-exam-v1","image_url":"https://handyai.news/api/benchmarks/agents-last-exam-v1.png","leader":{"model":"Claude Opus 5.5","score":63.2,"maker":"Anthropic"},"model_count":9,"drop_count":2,"first_date":"2026-06-26","last_date":"2026-10-26","frontier_gain":10.4,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":52.8,"maker":"openai"},{"date":"2026-09-03","model":"GPT-6 Astra","score":59.3,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":63.2,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":63.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"GPT-6.1 Terra","score":60,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":60,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":2},{"model":"GPT-6 Astra","score":59.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":3},{"model":"GPT-6.1 Sol","score":58.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":4},{"model":"GPT-6 Sol","score":56.4,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"GPT-6 Sol & GPT-6 Luna","slug":"gpt-6-sol-and-gpt-6-luna","url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","published":"2026-09-22"},"launch_score":56.4,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":5},{"model":"GPT-5.6 Sol","score":52.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6 Sol & GPT-6 Luna","slug":"gpt-6-sol-and-gpt-6-luna","url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","published":"2026-09-22"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":6},{"model":"Claude Opus 5","score":52.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6 Sol & GPT-6 Luna","slug":"gpt-6-sol-and-gpt-6-luna","url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","published":"2026-09-22"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":7},{"model":"GPT-6 Luna","score":50.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":8},{"model":"GPT-5.6 Terra","score":50.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":9}]},{"key":"chartography-with-tools","name":"Chartography (with tools)","about":null,"type":{"key":"multimodal","label":"Multimodal"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#chartography-with-tools","api_url":"https://handyai.news/api/benchmarks/chartography-with-tools","image_url":"https://handyai.news/api/benchmarks/chartography-with-tools.png","leader":{"model":"GPT-6 Astra","score":91,"maker":"OpenAI"},"model_count":9,"drop_count":2,"first_date":"2026-06-26","last_date":"2026-10-23","frontier_gain":11.1,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":79.9,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":84,"maker":"anthropic"},{"date":"2026-09-01","model":"Fable 5.1","score":88.4,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":91,"maker":"openai"}],"models":[{"model":"GPT-6 Astra","score":91,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":1},{"model":"Kimi K3.1 Pro","score":90.2,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":90.2,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":2},{"model":"Claude Opus 5.5","score":89,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":3},{"model":"Fable 5.1","score":88.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":4},{"model":"Claude Opus 5","score":84,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":5},{"model":"DeepSeek V4.1 Flash 1017","score":83,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":83,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":6},{"model":"GPT-5.6 Sol","score":79.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":7},{"model":"DeepSeek V4.1 Flash","score":78.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":8},{"model":"Kimi K3","score":68.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":9}]},{"key":"pinchbench","name":"PinchBench","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#pinchbench","api_url":"https://handyai.news/api/benchmarks/pinchbench","image_url":"https://handyai.news/api/benchmarks/pinchbench.png","leader":{"model":"Kimi K2.6","score":90.2,"maker":"Moonshot AI"},"model_count":9,"drop_count":2,"first_date":"2025-08-05","last_date":"2026-08-11","frontier_gain":33,"frontier":[{"date":"2025-08-05","model":"gpt-oss-20b","score":57.2,"maker":"openai"},{"date":"2025-12-15","model":"Nemotron 3 Nano","score":66.11,"maker":"nvidia"},{"date":"2026-03-27","model":"GLM-5.1","score":81.2,"maker":"zai"},{"date":"2026-04-22","model":"Kimi K2.6","score":90.2,"maker":"moonshot"}],"models":[{"model":"Kimi K2.6","score":90.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-04-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k26","rank":1},{"model":"Nemotron 3 Ultra","score":90,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":90,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":2},{"model":"Qwen3.6 35B A3B","score":88.07,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"first_cited","drop_url":null,"rank":3},{"model":"Qwen3.5 397B","score":86.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"first_cited","drop_url":null,"rank":4},{"model":"Nemotron 3.5 Lightning","score":85.37,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":85.37,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-lightning","rank":5},{"model":"GLM-5.1","score":81.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-03-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-51","rank":6},{"model":"Gemma 4 26B A4B","score":74.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":7},{"model":"Nemotron 3 Nano","score":66.11,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":8},{"model":"gpt-oss-20b","score":57.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":9}]},{"key":"agents-last-exam-cli","name":"Agents' Last Exam (CLI)","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#agents-last-exam-cli","api_url":"https://handyai.news/api/benchmarks/agents-last-exam-cli","image_url":"https://handyai.news/api/benchmarks/agents-last-exam-cli.png","leader":{"model":"Kimi K3.1 Pro","score":44.2,"maker":"Moonshot AI"},"model_count":8,"drop_count":2,"first_date":"2026-06-09","last_date":"2026-10-23","frontier_gain":20.4,"frontier":[{"date":"2026-06-09","model":"Fable 5","score":23.8,"maker":"anthropic"},{"date":"2026-06-26","model":"GPT-5.6 Sol","score":28.6,"maker":"openai"},{"date":"2026-09-03","model":"GPT-6 Astra","score":38.8,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":40.7,"maker":"anthropic"},{"date":"2026-10-23","model":"Kimi K3.1 Pro","score":44.2,"maker":"moonshot"}],"models":[{"model":"Kimi K3.1 Pro","score":44.2,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":44.2,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":1},{"model":"Claude Opus 5.5","score":40.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2},{"model":"GPT-6 Astra","score":38.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":3},{"model":"DeepSeek V4.1 Flash 1017","score":36,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":4},{"model":"GPT-5.6 Sol","score":28.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":5},{"model":"GLM-5.3","score":28.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":28.5,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":6},{"model":"Kimi K3","score":27.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":7},{"model":"Fable 5","score":23.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GLM-5.3","slug":"glm-53","url":"https://handyai.news/modeldrop/glm-53","published":"2026-08-14"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":8}]},{"key":"healthbench-professional","name":"HealthBench Professional","about":null,"type":{"key":"work","label":"Work"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#healthbench-professional","api_url":"https://handyai.news/api/benchmarks/healthbench-professional","image_url":"https://handyai.news/api/benchmarks/healthbench-professional.png","leader":{"model":"Mythos 5","score":66,"maker":"Anthropic"},"model_count":8,"drop_count":2,"first_date":"2026-04-07","last_date":"2026-09-21","frontier_gain":1.3,"frontier":[{"date":"2026-04-07","model":"Claude Mythos Preview","score":64.7,"maker":"anthropic"},{"date":"2026-06-09","model":"Mythos 5","score":66,"maker":"anthropic"}],"models":[{"model":"Mythos 5","score":66,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":66,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":1},{"model":"Claude Mythos Preview","score":64.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":2},{"model":"Fable 5.1","score":62.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":3},{"model":"GPT-5.6 Sol","score":60.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":4},{"model":"Claude Opus 4.8","score":56.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-48","rank":5},{"model":"Grok 4.7","score":56.7,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":56.7,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":6},{"model":"GPT-5.5","score":51.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-04-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-55","rank":7},{"model":"Grok 4.6","score":48.5,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":8}]},{"key":"osworld","name":"OSWorld","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#osworld","api_url":"https://handyai.news/api/benchmarks/osworld","image_url":"https://handyai.news/api/benchmarks/osworld.png","leader":{"model":"Claude Opus 5.5","score":81.8,"maker":"Anthropic"},"model_count":8,"drop_count":2,"first_date":"2025-05-22","last_date":"2026-10-23","frontier_gain":39.6,"frontier":[{"date":"2025-05-22","model":"Claude Sonnet 4","score":42.2,"maker":"anthropic"},{"date":"2025-08-05","model":"Claude Opus 4.1","score":44.4,"maker":"anthropic"},{"date":"2025-09-29","model":"Claude Sonnet 4.5","score":61.4,"maker":"anthropic"},{"date":"2026-09-01","model":"Fable 5.1","score":80.7,"maker":"anthropic"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":81.8,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":81.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"Fable 5.1","score":80.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":2},{"model":"Kimi K3.1 Pro","score":79.6,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":79.6,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":3},{"model":"GPT-6 Astra","score":75.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":4},{"model":"DeepSeek V4.1 Flash 1017","score":69.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":5},{"model":"Claude Sonnet 4.5","score":61.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":61.4,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":6},{"model":"Claude Opus 4.1","score":44.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":7},{"model":"Claude Sonnet 4","score":42.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Sonnet 4.5","slug":"claude-sonnet-45","url":"https://handyai.news/modeldrop/claude-sonnet-45","published":"2025-09-29"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":8}]},{"key":"swe-marathon","name":"SWE-Marathon","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#swe-marathon","api_url":"https://handyai.news/api/benchmarks/swe-marathon","image_url":"https://handyai.news/api/benchmarks/swe-marathon.png","leader":{"model":"Muse Wonder Full","score":57.5,"maker":"Meta"},"model_count":8,"drop_count":2,"first_date":"2026-08-14","last_date":"2026-10-23","frontier_gain":15,"frontier":[{"date":"2026-08-14","model":"GLM-5.3","score":42.5,"maker":"zai"},{"date":"2026-09-01","model":"Fable 5.1","score":45.6,"maker":"anthropic"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":46.3,"maker":"anthropic"},{"date":"2026-10-12","model":"Muse Wonder Full","score":57.5,"maker":"meta"}],"models":[{"model":"Muse Wonder Full","score":57.5,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":1},{"model":"Muse Wonder","score":53.1,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":53.1,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":2},{"model":"Kimi K3.1 Pro","score":51.7,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":51.7,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":3},{"model":"Claude Opus 5.5","score":46.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":4},{"model":"Fable 5.1","score":45.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":5},{"model":"GPT-6 Astra","score":42.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":6},{"model":"GLM-5.3","score":42.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":7},{"model":"DeepSeek V4.1 Flash 1017","score":40,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":8}]},{"key":"toolathlon-verified","name":"Toolathlon-Verified","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#toolathlon-verified","api_url":"https://handyai.news/api/benchmarks/toolathlon-verified","image_url":"https://handyai.news/api/benchmarks/toolathlon-verified.png","leader":{"model":"Claude Opus 5","score":80.6,"maker":"Anthropic"},"model_count":8,"drop_count":2,"first_date":"2026-04-02","last_date":"2026-10-24","frontier_gain":52.8,"frontier":[{"date":"2026-04-02","model":"Gemma 4","score":27.8,"maker":"google"},{"date":"2026-06-26","model":"GPT-5.6 Sol","score":74.9,"maker":"openai"},{"date":"2026-07-16","model":"Kimi K3","score":76.5,"maker":"moonshot"},{"date":"2026-07-24","model":"Claude Opus 5","score":80.6,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5","score":80.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":1},{"model":"MiMo-V2.6-Pro","score":76.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":76.9,"reports":1,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":2},{"model":"Kimi K3","score":76.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":3},{"model":"GPT-5.6 Sol","score":74.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiMo-V2.6-Pro","slug":"mimo-v26-pro","url":"https://handyai.news/modeldrop/mimo-v26-pro","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":4},{"model":"Nanbeige4.3-33B","score":70.4,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":70.4,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":5},{"model":"Qwen3.8","score":68.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":6},{"model":"Muse Glimmer-30B","score":47.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":7},{"model":"Gemma 4","score":27.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":8}]},{"key":"vals-index","name":"Vals Index","about":null,"type":{"key":"overall","label":"Overall index","folded":true},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#vals-index","api_url":"https://handyai.news/api/benchmarks/vals-index","image_url":"https://handyai.news/api/benchmarks/vals-index.png","leader":{"model":"Gemini 4 Argon","score":68.9,"maker":"Google DeepMind"},"model_count":8,"drop_count":2,"first_date":"2026-07-16","last_date":"2026-10-06","frontier_gain":18.6,"frontier":[{"date":"2026-07-16","model":"Kimi K3","score":50.3,"maker":"moonshot"},{"date":"2026-08-14","model":"GLM-5.3","score":53.51,"maker":"zai"},{"date":"2026-09-01","model":"Fable 5.1","score":65.8,"maker":"anthropic"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":67,"maker":"anthropic"},{"date":"2026-10-01","model":"Gemini 4 Argon","score":68.9,"maker":"google"}],"models":[{"model":"Gemini 4 Argon","score":68.9,"own_drop":true,"independent":true,"setting":"max","source":{"drop":"Gemini 4 Argon","slug":"gemini-4-argon","url":"https://handyai.news/modeldrop/gemini-4-argon","published":"2026-10-01"},"launch_score":68.9,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":1},{"model":"Claude Opus 5.5","score":67,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 4 Argon","slug":"gemini-4-argon","url":"https://handyai.news/modeldrop/gemini-4-argon","published":"2026-10-01"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2},{"model":"Fable 5.1","score":65.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 4 Argon","slug":"gemini-4-argon","url":"https://handyai.news/modeldrop/gemini-4-argon","published":"2026-10-01"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":3},{"model":"GPT-6 Astra","score":63.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 4 Argon","slug":"gemini-4-argon","url":"https://handyai.news/modeldrop/gemini-4-argon","published":"2026-10-01"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":4},{"model":"GLM-5.3","score":53.51,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":5},{"model":"Kimi K3","score":50.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":6},{"model":"Mistral Large 4","score":48.05,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":48.05,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":7},{"model":"DeepSeek V4 Pro 0813","score":47.63,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"first_cited","drop_url":null,"rank":8}]},{"key":"video-mmmu","name":"Video-MMMU","about":null,"type":{"key":"multimodal","label":"Multimodal"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#video-mmmu","api_url":"https://handyai.news/api/benchmarks/video-mmmu","image_url":"https://handyai.news/api/benchmarks/video-mmmu.png","leader":{"model":"Gemini 3 Pro","score":87.6,"maker":"Google DeepMind"},"model_count":8,"drop_count":2,"first_date":"2025-03-25","last_date":"2026-01-27","frontier_gain":4,"frontier":[{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":83.6,"maker":"google"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":87.6,"maker":"google"}],"models":[{"model":"Gemini 3 Pro","score":87.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":87.6,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":1},{"model":"Kimi K2.5","score":86.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":86.6,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":2},{"model":"GPT-5.2","score":85.9,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":3},{"model":"Claude Opus 4.5","score":84.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-11-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-45","rank":4},{"model":"Gemini 2.5 Pro","score":83.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":5},{"model":"GPT-5.1","score":80.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":6},{"model":"Qwen3-VL-235B-A22B-Thinking","score":80,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Kimi K2.5","slug":"kimi-k25","url":"https://handyai.news/modeldrop/kimi-k25","published":"2026-01-27"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"first_cited","drop_url":null,"rank":7},{"model":"Claude Sonnet 4.5","score":77.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":8}]},{"key":"bfcl-v3","name":"BFCL v3","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#bfcl-v3","api_url":"https://handyai.news/api/benchmarks/bfcl-v3","image_url":"https://handyai.news/api/benchmarks/bfcl-v3.png","leader":{"model":"Qwen3","score":70.8,"maker":"Alibaba"},"model_count":7,"drop_count":2,"first_date":"2024-12-05","last_date":"2025-08-18","frontier_gain":3,"frontier":[{"date":"2024-12-05","model":"o1","score":67.8,"maker":"openai"},{"date":"2025-04-29","model":"Qwen3","score":70.8,"maker":"qwen"}],"models":[{"model":"Qwen3","score":70.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":70.8,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-04-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3","rank":1},{"model":"o1","score":67.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":2},{"model":"Nemotron Nano 2","score":66.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":66.9,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-nano-2","rank":3},{"model":"Qwen3-8B","score":66.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"first_cited","drop_url":null,"rank":4},{"model":"o3-mini","score":64.6,"own_drop":false,"independent":false,"setting":"medium","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":5},{"model":"Gemini 2.5 Pro","score":62.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":6},{"model":"DeepSeek R1","score":56.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":7}]},{"key":"deepsearchqa","name":"DeepSearchQA","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#deepsearchqa","api_url":"https://handyai.news/api/benchmarks/deepsearchqa","image_url":"https://handyai.news/api/benchmarks/deepsearchqa.png","leader":{"model":"GPT-5.6 Sol","score":93.1,"maker":"OpenAI"},"model_count":7,"drop_count":2,"first_date":"2026-02-05","last_date":"2026-09-02","frontier_gain":19.4,"frontier":[{"date":"2026-02-05","model":"Claude Opus 4.6","score":73.7,"maker":"anthropic"},{"date":"2026-04-08","model":"Muse Spark","score":74.8,"maker":"meta"},{"date":"2026-06-26","model":"GPT-5.6 Sol","score":93.1,"maker":"openai"}],"models":[{"model":"GPT-5.6 Sol","score":93.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":1},{"model":"Claude Opus 5","score":90.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":2},{"model":"Muse Spark 1.3","score":90.3,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":90.3,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-13","rank":3},{"model":"Muse Spark","score":74.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":74.8,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-04-08","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark","rank":4},{"model":"Claude Opus 4.6","score":73.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":5},{"model":"GPT-5.4","score":73.6,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":6},{"model":"Gemini 3.1 Pro","score":69.7,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":7}]},{"key":"hle-verified","name":"HLE-Verified","about":null,"type":{"key":"reasoning","label":"Reasoning"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#hle-verified","api_url":"https://handyai.news/api/benchmarks/hle-verified","image_url":"https://handyai.news/api/benchmarks/hle-verified.png","leader":{"model":"Gemini 3.8 Flash","score":54.9,"maker":"Google DeepMind"},"model_count":7,"drop_count":2,"first_date":"2026-06-26","last_date":"2026-09-02","frontier_gain":0.4,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":54.5,"maker":"openai"},{"date":"2026-09-02","model":"Gemini 3.8 Flash","score":54.9,"maker":"google"}],"models":[{"model":"Gemini 3.8 Flash","score":54.9,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":54.9,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":1},{"model":"GPT-5.6 Sol","score":54.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":2},{"model":"Claude Opus 5","score":54.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":3},{"model":"Gemini 3.7 Flash","score":53.6,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":53.6,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-08-13","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-37-flash","rank":4},{"model":"Gemini 3.6 Flash","score":51.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3.7 Flash","slug":"gemini-37-flash","url":"https://handyai.news/modeldrop/gemini-37-flash","published":"2026-08-13"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":5},{"model":"GPT-5.6 Terra","score":51.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":6},{"model":"Claude Sonnet 5","score":31,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":7}]},{"key":"hmmt-2025","name":"HMMT 2025","about":null,"type":{"key":"math","label":"Math"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#hmmt-2025","api_url":"https://handyai.news/api/benchmarks/hmmt-2025","image_url":"https://handyai.news/api/benchmarks/hmmt-2025.png","leader":{"model":"Grok 4 Fast","score":93.3,"maker":"SpaceXAI"},"model_count":7,"drop_count":2,"first_date":"2025-03-25","last_date":"2025-09-19","frontier_gain":10.8,"frontier":[{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":82.5,"maker":"google"},{"date":"2025-07-09","model":"Grok 4","score":90,"maker":"xai"},{"date":"2025-08-07","model":"GPT-5","score":93.3,"maker":"openai"}],"models":[{"model":"Grok 4 Fast","score":93.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":93.3,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-09-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4-fast","rank":1},{"model":"GPT-5","score":93.3,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":2},{"model":"Grok 4","score":90,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":90,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":3},{"model":"GPT-5 mini","score":87.8,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4 Fast","slug":"grok-4-fast","url":"https://handyai.news/modeldrop/grok-4-fast","published":"2025-09-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-28","dated_by":"first_cited","drop_url":null,"rank":4},{"model":"Gemini 2.5 Pro","score":82.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4","slug":"grok-4","url":"https://handyai.news/modeldrop/grok-4","published":"2025-07-09"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":5},{"model":"o3","score":77.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4","slug":"grok-4","url":"https://handyai.news/modeldrop/grok-4","published":"2025-07-09"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":6},{"model":"Claude Opus 4","score":58.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4","slug":"grok-4","url":"https://handyai.news/modeldrop/grok-4","published":"2025-07-09"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":7}]},{"key":"simpleqa-verified","name":"SimpleQA Verified","about":null,"type":{"key":"knowledge","label":"Knowledge"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#simpleqa-verified","api_url":"https://handyai.news/api/benchmarks/simpleqa-verified","image_url":"https://handyai.news/api/benchmarks/simpleqa-verified.png","leader":{"model":"Gemini 3 Pro","score":72.1,"maker":"Google DeepMind"},"model_count":7,"drop_count":2,"first_date":"2025-03-25","last_date":"2025-12-17","frontier_gain":17.6,"frontier":[{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":54.5,"maker":"google"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":72.1,"maker":"google"}],"models":[{"model":"Gemini 3 Pro","score":72.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":72.1,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":1},{"model":"Gemini 3 Flash","score":68.7,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":68.7,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-flash","rank":2},{"model":"Gemini 2.5 Pro","score":54.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":3},{"model":"GPT-5.2","score":38,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":4},{"model":"GPT-5.1","score":34.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":5},{"model":"Claude Sonnet 4.5","score":29.3,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":6},{"model":"Grok 4.1 Fast","score":19.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 3 Flash","slug":"gemini-3-flash","url":"https://handyai.news/modeldrop/gemini-3-flash","published":"2025-12-17"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-12-17","dated_by":"first_cited","drop_url":null,"rank":7}]},{"key":"terminal-bench-2-1-standard-harness","name":"Terminal-Bench 2.1 (standard harness)","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#terminal-bench-2-1-standard-harness","api_url":"https://handyai.news/api/benchmarks/terminal-bench-2-1-standard-harness","image_url":"https://handyai.news/api/benchmarks/terminal-bench-2-1-standard-harness.png","leader":{"model":"Claude Opus 5","score":89.1,"maker":"Anthropic"},"model_count":7,"drop_count":2,"first_date":"2026-06-26","last_date":"2026-09-02","frontier_gain":1.1,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Terra","score":88,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":89.1,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5","score":89.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":1},{"model":"GPT-5.6 Terra","score":88,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Spark 1.2","slug":"muse-spark-12","url":"https://handyai.news/modeldrop/muse-spark-12","published":"2026-08-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":2},{"model":"GPT-5.6 Sol","score":88,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":3},{"model":"Gemini 3.8 Flash","score":87.6,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":4},{"model":"Muse Spark 1.3","score":85.4,"own_drop":true,"independent":true,"setting":"xhigh","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":85.4,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-13","rank":5},{"model":"Grok 4.5","score":81.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Muse Spark 1.2","slug":"muse-spark-12","url":"https://handyai.news/modeldrop/muse-spark-12","published":"2026-08-05"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":6},{"model":"Muse Spark 1.2","score":80.1,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Muse Spark 1.2","slug":"muse-spark-12","url":"https://handyai.news/modeldrop/muse-spark-12","published":"2026-08-05"},"launch_score":80.1,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-12","rank":7}]},{"key":"aa-briefcase","name":"AA-Briefcase","about":null,"type":{"key":"work","label":"Work"},"unit":"Elo","lower_is_better":false,"url":"https://handyai.news/benchmarks#aa-briefcase","api_url":"https://handyai.news/api/benchmarks/aa-briefcase","image_url":"https://handyai.news/api/benchmarks/aa-briefcase.png","leader":{"model":"Claude Opus 5","score":1715,"maker":"Anthropic"},"model_count":6,"drop_count":2,"first_date":"2026-06-09","last_date":"2026-10-27","frontier_gain":141,"frontier":[{"date":"2026-06-09","model":"Fable 5","score":1574,"maker":"anthropic"},{"date":"2026-07-24","model":"Claude Opus 5","score":1715,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5","score":1715,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":1},{"model":"Grok 4.6","score":1577,"own_drop":true,"independent":true,"setting":"high","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":1577,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":2},{"model":"Fable 5","score":1574,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":3},{"model":"Ling-3.1-flash-1019","score":1524,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Ling-3.1-flash-1019 model card","slug":"ling-31-flash-1019","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash-1019","published":"2026-10-27"},"launch_score":1524,"reports":1,"maker":{"key":"antgroup","name":"Ant Group","logo_url":"https://handyai.news/ui/makers/antgroup.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":4},{"model":"GPT-5.6 Sol","score":1502,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":5},{"model":"Grok 4.5","score":1313,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":6}]},{"key":"cursorbench","name":"CursorBench","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#cursorbench","api_url":"https://handyai.news/api/benchmarks/cursorbench","image_url":"https://handyai.news/api/benchmarks/cursorbench.png","leader":{"model":"GPT-5.4","score":63.9,"maker":"OpenAI"},"model_count":6,"drop_count":2,"first_date":"2025-10-29","last_date":"2026-03-19","frontier_gain":25.9,"frontier":[{"date":"2025-10-29","model":"Composer","score":38,"maker":"cursor"},{"date":"2026-02-05","model":"GPT-5.3-Codex","score":59.1,"maker":"openai"},{"date":"2026-03-05","model":"GPT-5.4","score":63.9,"maker":"openai"}],"models":[{"model":"GPT-5.4","score":63.9,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Composer 2","slug":"composer-2","url":"https://handyai.news/modeldrop/composer-2","published":"2026-03-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":1},{"model":"Composer 2","score":61.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Composer 2","slug":"composer-2","url":"https://handyai.news/modeldrop/composer-2","published":"2026-03-19"},"launch_score":61.3,"reports":1,"maker":{"key":"cursor","name":"Cursor","logo_url":"https://handyai.news/ui/makers/cursor.svg"},"open_weights":false,"released":"2026-03-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/composer-2","rank":2},{"model":"GPT-5.3-Codex","score":59.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Composer 2","slug":"composer-2","url":"https://handyai.news/modeldrop/composer-2","published":"2026-03-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-53-codex","rank":3},{"model":"Claude Opus 4.6","score":58.2,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Composer 2","slug":"composer-2","url":"https://handyai.news/modeldrop/composer-2","published":"2026-03-19"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":4},{"model":"Composer","score":38,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Composer","slug":"composer","url":"https://handyai.news/modeldrop/composer","published":"2025-10-29"},"launch_score":38,"reports":1,"maker":{"key":"cursor","name":"Cursor","logo_url":"https://handyai.news/ui/makers/cursor.svg"},"open_weights":false,"released":"2025-10-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/composer","rank":5},{"model":"Kimi K2.5","score":36,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Composer 2","slug":"composer-2","url":"https://handyai.news/modeldrop/composer-2","published":"2026-03-19"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-01-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k25","rank":6}]},{"key":"cursorbench-4-0","name":"CursorBench 4.0","about":"Long coding tasks inside Cursor.","type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#cursorbench-4-0","api_url":"https://handyai.news/api/benchmarks/cursorbench-4-0","image_url":"https://handyai.news/api/benchmarks/cursorbench-4-0.png","leader":{"model":"Claude Opus 5.5","score":57.8,"maker":"Anthropic"},"model_count":6,"drop_count":2,"first_date":"2026-06-26","last_date":"2026-09-22","frontier_gain":16.1,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":41.7,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":46.6,"maker":"anthropic"},{"date":"2026-09-01","model":"Fable 5.1","score":51.8,"maker":"anthropic"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":57.8,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":57.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Opus 5.5","slug":"claude-opus-55","url":"https://handyai.news/modeldrop/claude-opus-55","published":"2026-09-22"},"launch_score":57.8,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"Fable 5.1","score":51.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Claude Opus 5.5","slug":"claude-opus-55","url":"https://handyai.news/modeldrop/claude-opus-55","published":"2026-09-22"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":2},{"model":"Claude Opus 5","score":46.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 5.5","slug":"claude-opus-55","url":"https://handyai.news/modeldrop/claude-opus-55","published":"2026-09-22"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":3},{"model":"Grok 4.7","score":46.3,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":46.3,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":4},{"model":"GPT-5.6 Sol","score":41.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":5},{"model":"Grok 4.6","score":40.4,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":6}]},{"key":"multi-if","name":"Multi-IF","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#multi-if","api_url":"https://handyai.news/api/benchmarks/multi-if","image_url":"https://handyai.news/api/benchmarks/multi-if.png","leader":{"model":"LFM3-1.7B","score":89.1,"maker":"Liquid AI"},"model_count":6,"drop_count":2,"first_date":"2024-12-05","last_date":"2026-10-27","frontier_gain":40.3,"frontier":[{"date":"2024-12-05","model":"o1","score":48.8,"maker":"openai"},{"date":"2025-01-20","model":"DeepSeek R1","score":67.7,"maker":"deepseek"},{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":77.8,"maker":"google"},{"date":"2026-10-27","model":"LFM3-1.7B","score":89.1,"maker":"liquid"}],"models":[{"model":"LFM3-1.7B","score":89.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Liquid AI, \"Introducing LFM3\"","slug":"lfm3-17b","url":"https://www.liquid.ai/blog/lfm3","published":"2026-10-27"},"launch_score":89.1,"reports":1,"maker":{"key":"liquid","name":"Liquid AI","logo_url":"https://handyai.news/ui/makers/liquid.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":1},{"model":"Gemini 2.5 Pro","score":77.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":2},{"model":"Qwen3","score":71.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":71.9,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-04-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3","rank":3},{"model":"DeepSeek R1","score":67.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":4},{"model":"o1","score":48.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":5},{"model":"o3-mini","score":48.4,"own_drop":false,"independent":false,"setting":"medium","source":{"drop":"Qwen3","slug":"qwen3","url":"https://handyai.news/modeldrop/qwen3","published":"2025-04-29"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":6}]},{"key":"swe-atlas-qna","name":"SWE-Atlas-QnA","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#swe-atlas-qna","api_url":"https://handyai.news/api/benchmarks/swe-atlas-qna","image_url":"https://handyai.news/api/benchmarks/swe-atlas-qna.png","leader":{"model":"Fern 0.1","score":53.2,"maker":"Mirendil"},"model_count":6,"drop_count":2,"first_date":"2026-04-02","last_date":"2026-10-26","frontier_gain":49.9,"frontier":[{"date":"2026-04-02","model":"Gemma 4","score":3.3,"maker":"google"},{"date":"2026-06-04","model":"Nemotron 3 Ultra","score":16.1,"maker":"nvidia"},{"date":"2026-08-12","model":"Qwen3.8","score":41.9,"maker":"qwen"},{"date":"2026-10-24","model":"Nanbeige4.3-33B","score":50.8,"maker":"nanbeige"},{"date":"2026-10-26","model":"Fern 0.1","score":53.2,"maker":"mirendil"}],"models":[{"model":"Fern 0.1","score":53.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mirendil, \"Introducing Fern\"","slug":"fern-01","url":"https://mirendil.com/news/introducing-fern/","published":"2026-10-26"},"launch_score":53.2,"reports":1,"maker":{"key":"mirendil","name":"Mirendil","logo_url":"https://handyai.news/ui/makers/mirendil.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":null,"rank":1},{"model":"Nanbeige4.3-33B","score":50.8,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":50.8,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":2},{"model":"Qwen3.8","score":41.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":3},{"model":"Muse Glimmer-30B","score":17.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":4},{"model":"Nemotron 3 Ultra","score":16.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mirendil, \"Introducing Fern\"","slug":"fern-01","url":"https://mirendil.com/news/introducing-fern/","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":5},{"model":"Gemma 4","score":3.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":6}]},{"key":"tau-bench-airline","name":"TAU-bench Airline","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#tau-bench-airline","api_url":"https://handyai.news/api/benchmarks/tau-bench-airline","image_url":"https://handyai.news/api/benchmarks/tau-bench-airline.png","leader":{"model":"Claude Sonnet 4","score":60,"maker":"Anthropic"},"model_count":6,"drop_count":2,"first_date":"2025-02-24","last_date":"2025-08-05","frontier_gain":1.6,"frontier":[{"date":"2025-02-24","model":"Claude 3.7 Sonnet","score":58.4,"maker":"anthropic"},{"date":"2025-05-22","model":"Claude Sonnet 4","score":60,"maker":"anthropic"}],"models":[{"model":"Claude Sonnet 4","score":60,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":1},{"model":"Claude Opus 4","score":59.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":59.6,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":2},{"model":"Claude 3.7 Sonnet","score":58.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":3},{"model":"Claude Opus 4.1","score":56,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":56,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-41","rank":4},{"model":"o3","score":52,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4.1","slug":"claude-opus-41","url":"https://handyai.news/modeldrop/claude-opus-41","published":"2025-08-05"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":5},{"model":"GPT-4.1","score":49.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Opus 4 & Sonnet 4","slug":"claude-4","url":"https://handyai.news/modeldrop/claude-4","published":"2025-05-22"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":6}]},{"key":"artificial-analysis-coding-agent-index","name":"Artificial Analysis Coding Agent Index","about":null,"type":{"key":"coding","label":"Coding"},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#artificial-analysis-coding-agent-index","api_url":"https://handyai.news/api/benchmarks/artificial-analysis-coding-agent-index","image_url":"https://handyai.news/api/benchmarks/artificial-analysis-coding-agent-index.png","leader":{"model":"Claude Opus 5.5","score":62,"maker":"Anthropic"},"model_count":5,"drop_count":2,"first_date":"2026-08-12","last_date":"2026-10-28","frontier_gain":15,"frontier":[{"date":"2026-08-12","model":"Grok 4.6","score":47,"maker":"xai"},{"date":"2026-09-21","model":"Grok 4.7","score":56,"maker":"xai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":62,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":62,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"Gemini 4 Carbon","score":59,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":59,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":2},{"model":"Grok 4.7","score":56,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":56,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":3},{"model":"Gemini 4 Argon","score":53,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":4},{"model":"Grok 4.6","score":47,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":5}]},{"key":"enterpriseops-gym","name":"EnterpriseOps-Gym","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#enterpriseops-gym","api_url":"https://handyai.news/api/benchmarks/enterpriseops-gym","image_url":"https://handyai.news/api/benchmarks/enterpriseops-gym.png","leader":{"model":"Nanbeige4.3-33B","score":41,"maker":"Nanbeige"},"model_count":5,"drop_count":2,"first_date":"2026-04-02","last_date":"2026-10-27","frontier_gain":13.7,"frontier":[{"date":"2026-04-02","model":"Gemma 4","score":27.3,"maker":"google"},{"date":"2026-08-12","model":"Qwen3.8","score":40.2,"maker":"qwen"},{"date":"2026-10-24","model":"Nanbeige4.3-33B","score":41,"maker":"nanbeige"}],"models":[{"model":"Nanbeige4.3-33B","score":41,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":41,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":1},{"model":"Qwen3.8","score":40.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":2},{"model":"Muse Glimmer-30B","score":30.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":3},{"model":"LFM3-1.7B","score":28.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Liquid AI, \"Introducing LFM3\"","slug":"lfm3-17b","url":"https://www.liquid.ai/blog/lfm3","published":"2026-10-27"},"launch_score":28.2,"reports":1,"maker":{"key":"liquid","name":"Liquid AI","logo_url":"https://handyai.news/ui/makers/liquid.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":4},{"model":"Gemma 4","score":27.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":5}]},{"key":"frontiermath","name":"FrontierMath","about":null,"type":{"key":"math","label":"Math"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#frontiermath","api_url":"https://handyai.news/api/benchmarks/frontiermath","image_url":"https://handyai.news/api/benchmarks/frontiermath.png","leader":{"model":"GPT-5.1","score":26.7,"maker":"OpenAI"},"model_count":5,"drop_count":2,"first_date":"2024-12-05","last_date":"2025-11-12","frontier_gain":21.2,"frontier":[{"date":"2024-12-05","model":"o1","score":5.5,"maker":"openai"},{"date":"2025-01-20","model":"o1-mini","score":5.8,"maker":"openai"},{"date":"2025-01-31","model":"o3-mini","score":9.2,"maker":"openai"},{"date":"2025-08-07","model":"GPT-5","score":26.3,"maker":"openai"},{"date":"2025-11-12","model":"GPT-5.1","score":26.7,"maker":"openai"}],"models":[{"model":"GPT-5.1","score":26.7,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"GPT-5.1","slug":"gpt-51","url":"https://handyai.news/modeldrop/gpt-51","published":"2025-11-12"},"launch_score":26.7,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":1},{"model":"GPT-5","score":26.3,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.1","slug":"gpt-51","url":"https://handyai.news/modeldrop/gpt-51","published":"2025-11-12"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":2},{"model":"o3-mini","score":9.2,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":9.2,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":3},{"model":"o1-mini","score":5.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-20","dated_by":"first_cited","drop_url":null,"rank":4},{"model":"o1","score":5.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":5}]},{"key":"frontierswe-v2","name":"FrontierSWE v2","about":"Frontier software engineering tasks.","type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#frontierswe-v2","api_url":"https://handyai.news/api/benchmarks/frontierswe-v2","image_url":"https://handyai.news/api/benchmarks/frontierswe-v2.png","leader":{"model":"GPT-6 Astra","score":65.5,"maker":"OpenAI"},"model_count":5,"drop_count":2,"first_date":"2026-09-01","last_date":"2026-10-28","frontier_gain":9.2,"frontier":[{"date":"2026-09-01","model":"Fable 5.1","score":56.3,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":65.5,"maker":"openai"}],"models":[{"model":"GPT-6 Astra","score":65.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":1},{"model":"Claude Opus 5.5","score":62.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2},{"model":"Gemini 4 Carbon","score":61.2,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":61.2,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":3},{"model":"Fable 5.1","score":56.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":4},{"model":"Gemini 4 Argon","score":55,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":55,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":5}]},{"key":"professional-capture-the-flag-challenges","name":"Professional Capture-the-Flag Challenges","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#professional-capture-the-flag-challenges","api_url":"https://handyai.news/api/benchmarks/professional-capture-the-flag-challenges","image_url":"https://handyai.news/api/benchmarks/professional-capture-the-flag-challenges.png","leader":{"model":"GPT-5.2-Codex","score":88.1,"maker":"OpenAI"},"model_count":5,"drop_count":2,"first_date":"2025-09-15","last_date":"2025-12-18","frontier_gain":37.6,"frontier":[{"date":"2025-09-15","model":"GPT-5-Codex","score":50.5,"maker":"openai"},{"date":"2025-11-19","model":"GPT-5.1-Codex-Max","score":75.6,"maker":"openai"},{"date":"2025-12-11","model":"GPT-5.2","score":82.4,"maker":"openai"},{"date":"2025-12-18","model":"GPT-5.2-Codex","score":88.1,"maker":"openai"}],"models":[{"model":"GPT-5.2-Codex","score":88.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":88.1,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52-codex","rank":1},{"model":"GPT-5.2","score":82.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":2},{"model":"GPT-5.1-Codex-Max","score":75.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":76,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51-codex-max","rank":3},{"model":"GPT-5-Codex","score":50.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-09-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5-codex","rank":4},{"model":"GPT-5.1","score":43,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.1-Codex-Max","slug":"gpt-51-codex-max","url":"https://handyai.news/modeldrop/gpt-51-codex-max","published":"2025-11-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":5}]},{"key":"ruler-1m","name":"RULER @ 1M","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#ruler-1m","api_url":"https://handyai.news/api/benchmarks/ruler-1m","image_url":"https://handyai.news/api/benchmarks/ruler-1m.png","leader":{"model":"Nemotron 3 Ultra","score":94.7,"maker":"NVIDIA"},"model_count":5,"drop_count":2,"first_date":"2025-08-05","last_date":"2026-06-04","frontier_gain":72.4,"frontier":[{"date":"2025-08-05","model":"gpt-oss-120b","score":22.3,"maker":"openai"},{"date":"2026-03-11","model":"Nemotron 3 Super","score":91.75,"maker":"nvidia"},{"date":"2026-06-04","model":"Nemotron 3 Ultra","score":94.7,"maker":"nvidia"}],"models":[{"model":"Nemotron 3 Ultra","score":94.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":94.7,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":1},{"model":"Nemotron 3 Super","score":91.75,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":91.75,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-super","rank":2},{"model":"Qwen3.5-122B-A10B","score":91.33,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"first_cited","drop_url":null,"rank":3},{"model":"Qwen3.5 397B","score":90.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Ultra","slug":"nemotron-3-ultra","url":"https://handyai.news/modeldrop/nemotron-3-ultra","published":"2026-06-04"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"first_cited","drop_url":null,"rank":4},{"model":"gpt-oss-120b","score":22.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":5}]},{"key":"swe-lancer-ic-diamond","name":"SWE-Lancer IC Diamond","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#swe-lancer-ic-diamond","api_url":"https://handyai.news/api/benchmarks/swe-lancer-ic-diamond","image_url":"https://handyai.news/api/benchmarks/swe-lancer-ic-diamond.png","leader":{"model":"GPT-5.3-Codex","score":81.4,"maker":"OpenAI"},"model_count":5,"drop_count":2,"first_date":"2025-11-19","last_date":"2026-02-05","frontier_gain":1.5,"frontier":[{"date":"2025-11-19","model":"GPT-5.1-Codex-Max","score":79.9,"maker":"openai"},{"date":"2026-02-05","model":"GPT-5.3-Codex","score":81.4,"maker":"openai"}],"models":[{"model":"GPT-5.3-Codex","score":81.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.3-Codex","slug":"gpt-53-codex","url":"https://handyai.news/modeldrop/gpt-53-codex","published":"2026-02-05"},"launch_score":81.4,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-53-codex","rank":1},{"model":"GPT-5.1-Codex-Max","score":79.9,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.1-Codex-Max","slug":"gpt-51-codex-max","url":"https://handyai.news/modeldrop/gpt-51-codex-max","published":"2025-11-19"},"launch_score":79.9,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51-codex-max","rank":2},{"model":"GPT-5.2-Codex","score":76,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.3-Codex","slug":"gpt-53-codex","url":"https://handyai.news/modeldrop/gpt-53-codex","published":"2026-02-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52-codex","rank":3},{"model":"GPT-5.2","score":74.6,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.3-Codex","slug":"gpt-53-codex","url":"https://handyai.news/modeldrop/gpt-53-codex","published":"2026-02-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":4},{"model":"GPT-5.1-Codex","score":66.3,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.1-Codex-Max","slug":"gpt-51-codex-max","url":"https://handyai.news/modeldrop/gpt-51-codex-max","published":"2025-11-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-19","dated_by":"first_cited","drop_url":null,"rank":5}]},{"key":"mrcr-1m","name":"MRCR (1M)","about":null,"type":{"key":"long-context","label":"Long context"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#mrcr-1m","api_url":"https://handyai.news/api/benchmarks/mrcr-1m","image_url":"https://handyai.news/api/benchmarks/mrcr-1m.png","leader":{"model":"Gemini 1.5 Pro","score":82.6,"maker":"Google DeepMind"},"model_count":4,"drop_count":2,"first_date":"2024-12-05","last_date":"2025-04-17","frontier_gain":0,"frontier":[{"date":"2024-12-05","model":"Gemini 1.5 Pro","score":82.6,"maker":"google"}],"models":[{"model":"Gemini 1.5 Pro","score":82.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":1},{"model":"Gemini 1.5 Flash","score":71.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"first_cited","drop_url":null,"rank":2},{"model":"Gemini 2.5 Flash","score":66.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":66.3,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-04-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-flash","rank":3},{"model":"Gemini 2.0 Flash","score":48.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.5 Flash","slug":"gemini-25-flash","url":"https://handyai.news/modeldrop/gemini-25-flash","published":"2025-04-17"},"launch_score":69.2,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":4}]},{"key":"paperbench-10","name":"PaperBench-10","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#paperbench-10","api_url":"https://handyai.news/api/benchmarks/paperbench-10","image_url":"https://handyai.news/api/benchmarks/paperbench-10.png","leader":{"model":"GPT-5.2-Codex","score":43,"maker":"OpenAI"},"model_count":4,"drop_count":2,"first_date":"2025-11-12","last_date":"2025-12-18","frontier_gain":9,"frontier":[{"date":"2025-11-12","model":"GPT-5.1","score":34,"maker":"openai"},{"date":"2025-11-19","model":"GPT-5.1-Codex-Max","score":40,"maker":"openai"},{"date":"2025-12-18","model":"GPT-5.2-Codex","score":43,"maker":"openai"}],"models":[{"model":"GPT-5.2-Codex","score":43,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":43,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52-codex","rank":1},{"model":"GPT-5.1-Codex-Max","score":40,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":40,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51-codex-max","rank":2},{"model":"GPT-5.2","score":39,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":3},{"model":"GPT-5.1","score":34,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.1-Codex-Max","slug":"gpt-51-codex-max","url":"https://handyai.news/modeldrop/gpt-51-codex-max","published":"2025-11-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":4}]},{"key":"cybench","name":"CyBench","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#cybench","api_url":"https://handyai.news/api/benchmarks/cybench","image_url":"https://handyai.news/api/benchmarks/cybench.png","leader":{"model":"Grok 4","score":43,"maker":"SpaceXAI"},"model_count":3,"drop_count":2,"first_date":"2025-07-09","last_date":"2025-11-17","frontier_gain":0,"frontier":[{"date":"2025-07-09","model":"Grok 4","score":43,"maker":"xai"}],"models":[{"model":"Grok 4","score":43,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":1},{"model":"Grok 4.1","score":39,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":39,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-11-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-41","rank":2},{"model":"Grok Code Fast 1","score":22.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok Code Fast 1","slug":"grok-code-fast-1","url":"https://handyai.news/modeldrop/grok-code-fast-1","published":"2025-08-28"},"launch_score":22.5,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-08-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-code-fast-1","rank":3}]},{"key":"livebench-coding","name":"LiveBench Coding","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#livebench-coding","api_url":"https://handyai.news/api/benchmarks/livebench-coding","image_url":"https://handyai.news/api/benchmarks/livebench-coding.png","leader":{"model":"o3-mini","score":84.6,"maker":"OpenAI"},"model_count":3,"drop_count":2,"first_date":"2024-12-05","last_date":"2025-01-31","frontier_gain":17.2,"frontier":[{"date":"2024-12-05","model":"o1","score":67.4,"maker":"openai"},{"date":"2025-01-31","model":"o3-mini","score":84.6,"maker":"openai"}],"models":[{"model":"o3-mini","score":84.6,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":84.6,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":1},{"model":"o1","score":67.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3-mini","slug":"o3-mini","url":"https://handyai.news/modeldrop/o3-mini","published":"2025-01-31"},"launch_score":76.6,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":2},{"model":"o1-preview","score":52.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o1","slug":"o1","url":"https://handyai.news/modeldrop/o1","published":"2024-12-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":3}]},{"key":"mask-dishonesty-rate","name":"MASK dishonesty rate","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":true,"url":"https://handyai.news/benchmarks#mask-dishonesty-rate","api_url":"https://handyai.news/api/benchmarks/mask-dishonesty-rate","image_url":"https://handyai.news/api/benchmarks/mask-dishonesty-rate.png","leader":{"model":"Grok 4","score":43,"maker":"SpaceXAI"},"model_count":3,"drop_count":2,"first_date":"2025-07-09","last_date":"2025-11-17","frontier_gain":0,"frontier":[{"date":"2025-07-09","model":"Grok 4","score":43,"maker":"xai"}],"models":[{"model":"Grok 4","score":43,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":1},{"model":"Grok 4.1","score":49,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":49,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-11-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-41","rank":2},{"model":"Grok Code Fast 1","score":71.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok Code Fast 1","slug":"grok-code-fast-1","url":"https://handyai.news/modeldrop/grok-code-fast-1","published":"2025-08-28"},"launch_score":71.9,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-08-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-code-fast-1","rank":3}]},{"key":"frontiercode-2-0","name":"FrontierCode 2.0","about":"Real coding tasks in open-source repos (Cognition), run by Artificial Analysis.","type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#frontiercode-2-0","api_url":"https://handyai.news/api/benchmarks/frontiercode-2-0","image_url":"https://handyai.news/api/benchmarks/frontiercode-2-0.png","leader":{"model":"Claude Opus 5.5","score":46.4,"maker":"Anthropic"},"model_count":27,"drop_count":1,"first_date":"2026-07-15","last_date":"2026-11-04","frontier_gain":42.5,"frontier":[{"date":"2026-07-15","model":"Inkling","score":3.9,"maker":"thinkingmachines"},{"date":"2026-07-19","model":"Qwen3.8-Max","score":23.4,"maker":"qwen"},{"date":"2026-09-01","model":"Fable 5.1","score":31.6,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":42.4,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":46.4,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":46.4,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"GPT-6 Astra","score":42.4,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":2},{"model":"GPT-6.1 Terra","score":42.3,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":3},{"model":"Claude Sonnet 5.5","score":41.7,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":4},{"model":"Gemini 4 Carbon","score":40.6,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":5},{"model":"GPT-6.1 Sol","score":40.6,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":6},{"model":"Muse Wonder Full","score":37.8,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":7},{"model":"Claude Haiku 5.5","score":34.9,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":8},{"model":"Fable 5.1","score":31.6,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":9},{"model":"Hy4","score":31,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Tencent Hunyuan Hy4 launch post","slug":"hy4","url":"https://x.com/TencentHunyuan/status/2117870250838630539","published":"2026-11-04"},"launch_score":31,"reports":1,"maker":{"key":"tencent","name":"Tencent","logo_url":"https://handyai.news/ui/makers/tencent.svg"},"open_weights":true,"released":"2026-11-04","dated_by":"own_drop","drop_url":null,"rank":10},{"model":"Grok 4.7","score":28.7,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":11},{"model":"DeepSeek V4.1 Flash 1017","score":28.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":12},{"model":"Muse Wonder","score":27.9,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":13},{"model":"Gemini 4 Argon","score":25.7,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":14},{"model":"Kimi K3.1 Pro","score":25.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":15},{"model":"GPT-6 Luna","score":25.2,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":16},{"model":"MiMo-V2.6-Pro","score":24,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":17},{"model":"Qwen3.8-Max","score":23.4,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":18},{"model":"GLM-5.3","score":21.6,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":19},{"model":"MiniMax M3.1 Flash","score":20.8,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":20},{"model":"Gemini 3.8 Flash","score":18.7,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":21},{"model":"Ling-3.1-flash-1019","score":17,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"antgroup","name":"Ant Group","logo_url":"https://handyai.news/ui/makers/antgroup.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":22},{"model":"Mistral Large 4","score":15,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":23},{"model":"Qwen3.8","score":14.9,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":24},{"model":"Muse Spark Open","score":7.6,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":25},{"model":"Nemotron 3.5 Super VL","score":6.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":26},{"model":"Inkling","score":3.9,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":27}]},{"key":"artificial-analysis-intelligence-index-v5-0","name":"Artificial Analysis Intelligence Index v5.0","about":null,"type":{"key":"overall","label":"Overall index","folded":true},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#artificial-analysis-intelligence-index-v5-0","api_url":"https://handyai.news/api/benchmarks/artificial-analysis-intelligence-index-v5-0","image_url":"https://handyai.news/api/benchmarks/artificial-analysis-intelligence-index-v5-0.png","leader":{"model":"Gemini 4 Carbon","score":60,"maker":"Google DeepMind"},"model_count":26,"drop_count":1,"first_date":"2026-07-15","last_date":"2026-10-28","frontier_gain":47.5,"frontier":[{"date":"2026-07-15","model":"Inkling","score":12.5,"maker":"thinkingmachines"},{"date":"2026-07-19","model":"Qwen3.8-Max","score":33.7,"maker":"qwen"},{"date":"2026-09-01","model":"Fable 5.1","score":49.2,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":56.9,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":59.1,"maker":"anthropic"},{"date":"2026-10-28","model":"Gemini 4 Carbon","score":60,"maker":"google"}],"models":[{"model":"Gemini 4 Carbon","score":60,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":59,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":1},{"model":"Claude Opus 5.5","score":59.1,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2},{"model":"GPT-6 Astra","score":56.9,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":3},{"model":"GPT-6.1 Sol","score":54.6,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":4},{"model":"Claude Sonnet 5.5","score":54.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":5},{"model":"GPT-6.1 Terra","score":51.3,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":6},{"model":"Gemini 4 Argon","score":50.4,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":7},{"model":"Muse Wonder Full","score":50.1,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":8},{"model":"Fable 5.1","score":49.2,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":9},{"model":"Kimi K3.1 Pro","score":46.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":10},{"model":"Muse Wonder","score":43.6,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":11},{"model":"Claude Haiku 5.5","score":36.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":12},{"model":"Grok 4.7","score":35.5,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":13},{"model":"Qwen3.8-Max","score":33.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":14},{"model":"DeepSeek V4.1 Flash 1017","score":32.6,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":15},{"model":"GLM-5.3","score":31.7,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":16},{"model":"Gemini 3.8 Flash","score":30.8,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":17},{"model":"MiMo-V2.6-Pro","score":29.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":18},{"model":"Ling-3.1-flash-1019","score":29.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"antgroup","name":"Ant Group","logo_url":"https://handyai.news/ui/makers/antgroup.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":19},{"model":"MiniMax M3.1 Flash","score":29.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":20},{"model":"GPT-6 Luna","score":27.9,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":21},{"model":"Mistral Large 4","score":26.1,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":22},{"model":"Muse Spark Open","score":24.9,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":23},{"model":"Qwen3.8","score":20.1,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":24},{"model":"Nemotron 3.5 Super VL","score":13.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":25},{"model":"Inkling","score":12.5,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":1,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":26}]},{"key":"vals-index-v2-2","name":"Vals Index v2.2","about":null,"type":{"key":"overall","label":"Overall index","folded":true},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#vals-index-v2-2","api_url":"https://handyai.news/api/benchmarks/vals-index-v2-2","image_url":"https://handyai.news/api/benchmarks/vals-index-v2-2.png","leader":{"model":"Claude Sonnet 5.5","score":70.85,"maker":"Anthropic"},"model_count":24,"drop_count":1,"first_date":"2026-07-15","last_date":"2026-10-28","frontier_gain":41.9,"frontier":[{"date":"2026-07-15","model":"Inkling","score":28.96,"maker":"thinkingmachines"},{"date":"2026-07-16","model":"Kimi K3","score":52.95,"maker":"moonshot"},{"date":"2026-07-19","model":"Qwen3.8-Max","score":59.05,"maker":"qwen"},{"date":"2026-08-14","model":"GLM-5.3","score":59.73,"maker":"zai"},{"date":"2026-09-03","model":"GPT-6 Astra","score":65.02,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":70.18,"maker":"anthropic"},{"date":"2026-09-28","model":"Claude Sonnet 5.5","score":70.85,"maker":"anthropic"}],"models":[{"model":"Claude Sonnet 5.5","score":70.85,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":1},{"model":"Gemini 4 Carbon","score":70.45,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":70.4,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":2},{"model":"Claude Opus 5.5","score":70.18,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":3},{"model":"Gemini 4 Argon","score":69.24,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":4},{"model":"Muse Wonder Full","score":66.79,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":5},{"model":"GPT-6 Astra","score":65.02,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":6},{"model":"Muse Wonder","score":64.88,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":7},{"model":"GPT-6.1 Sol","score":63.94,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":8},{"model":"Kimi K3.1 Pro","score":62.52,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":9},{"model":"DeepSeek V4.1 Flash 1017","score":61.26,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":10},{"model":"GPT-6.1 Terra","score":60.54,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":11},{"model":"GLM-5.3","score":59.73,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":12},{"model":"Qwen3.8-Max","score":59.05,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":13},{"model":"Claude Haiku 5.5","score":58.73,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":14},{"model":"Grok 4.7","score":58.56,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":15},{"model":"MiMo-V2.6-Pro","score":57.81,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":16},{"model":"Mistral Large 4","score":56.53,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":17},{"model":"Gemini 3.8 Flash","score":55.95,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":18},{"model":"MiniMax M3.1 Flash","score":53.78,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":19},{"model":"Kimi K3","score":52.95,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":20},{"model":"GPT-6 Luna","score":52.32,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":21},{"model":"Muse Spark Open","score":48.57,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":22},{"model":"Nemotron 3.5 Super VL","score":29.76,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":23},{"model":"Inkling","score":28.96,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Vals Index v2.2 (2026-11-04)","slug":"vals-auto-2026-11-04","url":"https://www.vals.ai/benchmarks/vals_index","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":24}]},{"key":"artificial-analysis-cyber-index","name":"Artificial Analysis Cyber Index","about":null,"type":{"key":"cyber","label":"Cyber"},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#artificial-analysis-cyber-index","api_url":"https://handyai.news/api/benchmarks/artificial-analysis-cyber-index","image_url":"https://handyai.news/api/benchmarks/artificial-analysis-cyber-index.png","leader":{"model":"Kimi K3.1 Pro","score":64.1,"maker":"Moonshot AI"},"model_count":14,"drop_count":1,"first_date":"2026-07-16","last_date":"2026-10-23","frontier_gain":22.8,"frontier":[{"date":"2026-07-16","model":"Kimi K3","score":41.3,"maker":"moonshot"},{"date":"2026-09-21","model":"Grok 4.7","score":56.4,"maker":"xai"},{"date":"2026-10-17","model":"DeepSeek V4.1 Flash 1017","score":56.8,"maker":"deepseek"},{"date":"2026-10-23","model":"Kimi K3.1 Pro","score":64.1,"maker":"moonshot"}],"models":[{"model":"Kimi K3.1 Pro","score":64.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":1},{"model":"DeepSeek V4.1 Flash 1017","score":56.8,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":2},{"model":"Grok 4.7","score":56.4,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":3},{"model":"MiMo-V2.6-Pro","score":56.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":4},{"model":"GPT-6 Luna","score":52.7,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":5},{"model":"Mistral Large 4","score":50.9,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":50,"reports":3,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":6},{"model":"GLM-5.3-Flash","score":50,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"first_cited","drop_url":null,"rank":7},{"model":"Kimi K3","score":41.3,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":8},{"model":"GLM-5.3","score":36.4,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":9},{"model":"GPT-6 Astra","score":33.4,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":10},{"model":"Claude Opus 5.5","score":28.7,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":3,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":11},{"model":"Gemini 3.8 Flash","score":25.1,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":12},{"model":"Fable 5.1","score":25,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":13},{"model":"Qwen3.8","score":12.7,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":14}]},{"key":"apex-bench-v1-1","name":"Apex Bench v1.1","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#apex-bench-v1-1","api_url":"https://handyai.news/api/benchmarks/apex-bench-v1-1","image_url":"https://handyai.news/api/benchmarks/apex-bench-v1-1.png","leader":{"model":"DeepSeek V4.1 Flash 1017","score":72.1,"maker":"DeepSeek"},"model_count":6,"drop_count":1,"first_date":"2026-06-26","last_date":"2026-10-17","frontier_gain":17.8,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":54.3,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":65.8,"maker":"anthropic"},{"date":"2026-10-17","model":"DeepSeek V4.1 Flash 1017","score":72.1,"maker":"deepseek"}],"models":[{"model":"DeepSeek V4.1 Flash 1017","score":72.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":72.1,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":1},{"model":"Claude Opus 5","score":65.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":2},{"model":"DeepSeek V4.1 Flash","score":64.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":3},{"model":"GLM-5.3","score":56.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":4},{"model":"GPT-5.6 Sol","score":54.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":5},{"model":"Kimi K3","score":47.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"DeepSeek V4.1 Flash 1017","slug":"deepseek-v41-flash-1017","url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","published":"2026-10-17"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":6}]},{"key":"bankertoolbench","name":"BankerToolBench","about":null,"type":{"key":"other","label":"Other"},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#bankertoolbench","api_url":"https://handyai.news/api/benchmarks/bankertoolbench","image_url":"https://handyai.news/api/benchmarks/bankertoolbench.png","leader":{"model":"MiniMax M3.1 Flash","score":76.2,"maker":"MiniMax"},"model_count":6,"drop_count":1,"first_date":"2026-06-26","last_date":"2026-10-19","frontier_gain":1.6,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":74.6,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":76.2,"maker":"anthropic"}],"models":[{"model":"MiniMax M3.1 Flash","score":76.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":76.2,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":1},{"model":"Claude Opus 5","score":76.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":2},{"model":"GPT-5.6 Sol","score":74.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":3},{"model":"GLM-5.3","score":74.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":4},{"model":"DeepSeek V4.1 Flash","score":73.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":5},{"model":"Kimi K3","score":69.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax M3.1 Flash","slug":"minimax-m31-flash","url":"https://handyai.news/modeldrop/minimax-m31-flash","published":"2026-10-19"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":6}]},{"key":"frontier-finance","name":"Frontier Finance","about":null,"type":{"key":"work","label":"Work"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#frontier-finance","api_url":"https://handyai.news/api/benchmarks/frontier-finance","image_url":"https://handyai.news/api/benchmarks/frontier-finance.png","leader":{"model":"Fable 5.1","score":60.8,"maker":"Anthropic"},"model_count":6,"drop_count":1,"first_date":"2026-08-14","last_date":"2026-10-23","frontier_gain":3.3,"frontier":[{"date":"2026-08-14","model":"GLM-5.3","score":57.5,"maker":"zai"},{"date":"2026-09-01","model":"Fable 5.1","score":60.8,"maker":"anthropic"}],"models":[{"model":"Fable 5.1","score":60.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":1},{"model":"Kimi K3.1 Pro","score":58.6,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":58.6,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":2},{"model":"GLM-5.3","score":57.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":3},{"model":"Claude Opus 5.5","score":56.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":4},{"model":"DeepSeek V4.1 Flash 1017","score":54.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":5},{"model":"GPT-6 Astra","score":45,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":6}]},{"key":"officeqa-pro-v2","name":"OfficeQA Pro v2","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#officeqa-pro-v2","api_url":"https://handyai.news/api/benchmarks/officeqa-pro-v2","image_url":"https://handyai.news/api/benchmarks/officeqa-pro-v2.png","leader":{"model":"Kimi K3.1 Pro","score":64.4,"maker":"Moonshot AI"},"model_count":6,"drop_count":1,"first_date":"2026-08-14","last_date":"2026-10-23","frontier_gain":33.3,"frontier":[{"date":"2026-08-14","model":"GLM-5.3","score":31.1,"maker":"zai"},{"date":"2026-09-01","model":"Fable 5.1","score":54.5,"maker":"anthropic"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":62.2,"maker":"anthropic"},{"date":"2026-10-23","model":"Kimi K3.1 Pro","score":64.4,"maker":"moonshot"}],"models":[{"model":"Kimi K3.1 Pro","score":64.4,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":64.4,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":1},{"model":"Claude Opus 5.5","score":62.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2},{"model":"Fable 5.1","score":54.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":3},{"model":"GPT-6 Astra","score":54.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":4},{"model":"DeepSeek V4.1 Flash 1017","score":40.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":5},{"model":"GLM-5.3","score":31.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":6}]},{"key":"tokens-per-second","name":"Tokens per second","about":null,"type":{"key":"other","label":"Other"},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#tokens-per-second","api_url":"https://handyai.news/api/benchmarks/tokens-per-second","image_url":"https://handyai.news/api/benchmarks/tokens-per-second.png","leader":{"model":"Grok Code Fast 1","score":160,"maker":"SpaceXAI"},"model_count":6,"drop_count":1,"first_date":"2025-03-25","last_date":"2025-08-28","frontier_gain":67.6,"frontier":[{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":92.39,"maker":"google"},{"date":"2025-08-28","model":"Grok Code Fast 1","score":160,"maker":"xai"}],"models":[{"model":"Grok Code Fast 1","score":160,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok Code Fast 1","slug":"grok-code-fast-1","url":"https://handyai.news/modeldrop/grok-code-fast-1","published":"2025-08-28"},"launch_score":160,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-08-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-code-fast-1","rank":1},{"model":"Gemini 2.5 Pro","score":92.39,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok Code Fast 1","slug":"grok-code-fast-1","url":"https://handyai.news/modeldrop/grok-code-fast-1","published":"2025-08-28"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":2},{"model":"Qwen3-Coder","score":80,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok Code Fast 1","slug":"grok-code-fast-1","url":"https://handyai.news/modeldrop/grok-code-fast-1","published":"2025-08-28"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-08-28","dated_by":"first_cited","drop_url":null,"rank":3},{"model":"Claude Sonnet 4","score":78.71,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok Code Fast 1","slug":"grok-code-fast-1","url":"https://handyai.news/modeldrop/grok-code-fast-1","published":"2025-08-28"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":4},{"model":"GPT-5","score":50.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok Code Fast 1","slug":"grok-code-fast-1","url":"https://handyai.news/modeldrop/grok-code-fast-1","published":"2025-08-28"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":5},{"model":"Grok 4","score":40.63,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok Code Fast 1","slug":"grok-code-fast-1","url":"https://handyai.news/modeldrop/grok-code-fast-1","published":"2025-08-28"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":6}]},{"key":"artificial-analysis-coding-agent-index-v2-0","name":"Artificial Analysis Coding Agent Index v2.0","about":null,"type":{"key":"coding","label":"Coding"},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#artificial-analysis-coding-agent-index-v2-0","api_url":"https://handyai.news/api/benchmarks/artificial-analysis-coding-agent-index-v2-0","image_url":"https://handyai.news/api/benchmarks/artificial-analysis-coding-agent-index-v2-0.png","leader":{"model":"Claude Opus 5.5","score":62,"maker":"Anthropic"},"model_count":5,"drop_count":1,"first_date":"2026-09-22","last_date":"2026-10-26","frontier_gain":0,"frontier":[{"date":"2026-09-22","model":"Claude Opus 5.5","score":62,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":62,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"Claude Sonnet 5.5","score":59,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":2},{"model":"GPT-6.1 Sol","score":54,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":3},{"model":"GPT-6.1 Terra","score":48,"own_drop":true,"independent":true,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":48,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":4},{"model":"GPT-6 Luna","score":31,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":5}]},{"key":"benchcad-with-tools","name":"BenchCAD (with tools)","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#benchcad-with-tools","api_url":"https://handyai.news/api/benchmarks/benchcad-with-tools","image_url":"https://handyai.news/api/benchmarks/benchcad-with-tools.png","leader":{"model":"Claude Opus 5.5","score":96.2,"maker":"Anthropic"},"model_count":5,"drop_count":1,"first_date":"2026-09-01","last_date":"2026-10-23","frontier_gain":3.6,"frontier":[{"date":"2026-09-01","model":"Fable 5.1","score":92.6,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":95.9,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":96.2,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":96.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"GPT-6 Astra","score":95.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":2},{"model":"Kimi K3.1 Pro","score":92.7,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":92.7,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":3},{"model":"Fable 5.1","score":92.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":4},{"model":"DeepSeek V4.1 Flash 1017","score":81.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Kimi K3.1 Pro","slug":"kimi-k31-pro","url":"https://handyai.news/modeldrop/kimi-k31-pro","published":"2026-10-23"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":5}]},{"key":"cybergym-e2e","name":"CyberGym-E2E","about":null,"type":{"key":"cyber","label":"Cyber"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#cybergym-e2e","api_url":"https://handyai.news/api/benchmarks/cybergym-e2e","image_url":"https://handyai.news/api/benchmarks/cybergym-e2e.png","leader":{"model":"Mistral Large 4","score":82,"maker":"Mistral AI"},"model_count":5,"drop_count":1,"first_date":"2026-07-16","last_date":"2026-10-06","frontier_gain":24,"frontier":[{"date":"2026-07-16","model":"Kimi K3","score":58,"maker":"moonshot"},{"date":"2026-09-21","model":"MiMo-V2.6-Pro","score":79,"maker":"xiaomi"},{"date":"2026-10-06","model":"Mistral Large 4","score":82,"maker":"mistral"}],"models":[{"model":"Mistral Large 4","score":82,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":82,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":1},{"model":"MiMo-V2.6-Pro","score":79,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":2},{"model":"Grok 4.7","score":74,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":3},{"model":"Kimi K3","score":58,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":4},{"model":"GLM-5.3","score":29,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":5}]},{"key":"eq-bench3","name":"EQ-Bench3","about":null,"type":{"key":"other","label":"Other"},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#eq-bench3","api_url":"https://handyai.news/api/benchmarks/eq-bench3","image_url":"https://handyai.news/api/benchmarks/eq-bench3.png","leader":{"model":"Grok 4.1","score":1586,"maker":"SpaceXAI"},"model_count":5,"drop_count":1,"first_date":"2025-03-25","last_date":"2025-11-17","frontier_gain":126,"frontier":[{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":1460,"maker":"google"},{"date":"2025-07-11","model":"Kimi K2","score":1561,"maker":"moonshot"},{"date":"2025-11-17","model":"Grok 4.1","score":1586,"maker":"xai"}],"models":[{"model":"Grok 4.1","score":1586,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":1586,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-11-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-41","rank":1},{"model":"Kimi K2","score":1561,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-07-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k2","rank":2},{"model":"Gemini 2.5 Pro","score":1460,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":3},{"model":"Claude Opus 4","score":1304,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":4},{"model":"Grok 4","score":1206,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":5}]},{"key":"factual-error-rate","name":"Factual error rate","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":true,"url":"https://handyai.news/benchmarks#factual-error-rate","api_url":"https://handyai.news/api/benchmarks/factual-error-rate","image_url":"https://handyai.news/api/benchmarks/factual-error-rate.png","leader":{"model":"GPT-6 Astra","score":4.2,"maker":"OpenAI"},"model_count":5,"drop_count":1,"first_date":"2026-06-26","last_date":"2026-10-26","frontier_gain":18.1,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Terra","score":22.3,"maker":"openai"},{"date":"2026-09-03","model":"GPT-6 Astra","score":4.2,"maker":"openai"}],"models":[{"model":"GPT-6 Astra","score":4.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":1},{"model":"GPT-6.1 Terra","score":4.3,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":4.3,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":2},{"model":"GPT-6.1 Sol","score":4.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":3},{"model":"GPT-6 Luna","score":18.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":4},{"model":"GPT-5.6 Terra","score":22.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":5}]},{"key":"ifbench-loose","name":"IFBench (loose)","about":null,"type":{"key":"instructions","label":"Instructions"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#ifbench-loose","api_url":"https://handyai.news/api/benchmarks/ifbench-loose","image_url":"https://handyai.news/api/benchmarks/ifbench-loose.png","leader":{"model":"Gemma 4 26B A4B","score":77.25,"maker":"Google DeepMind"},"model_count":5,"drop_count":1,"first_date":"2025-08-05","last_date":"2026-08-11","frontier_gain":8.8,"frontier":[{"date":"2025-08-05","model":"gpt-oss-20b","score":68.5,"maker":"openai"},{"date":"2025-12-15","model":"Nemotron 3 Nano","score":72.17,"maker":"nvidia"},{"date":"2026-04-02","model":"Gemma 4 26B A4B","score":77.25,"maker":"google"}],"models":[{"model":"Gemma 4 26B A4B","score":77.25,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":1},{"model":"Nemotron 3 Nano","score":72.17,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":2},{"model":"Nemotron 3.5 Lightning","score":71.88,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":71.88,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-lightning","rank":3},{"model":"gpt-oss-20b","score":68.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":4},{"model":"Qwen3.6 35B A3B","score":63.71,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Lightning","slug":"nemotron-35-lightning","url":"https://handyai.news/modeldrop/nemotron-35-lightning","published":"2026-08-11"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-11","dated_by":"first_cited","drop_url":null,"rank":5}]},{"key":"memorybench","name":"MemoryBench","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#memorybench","api_url":"https://handyai.news/api/benchmarks/memorybench","image_url":"https://handyai.news/api/benchmarks/memorybench.png","leader":{"model":"Muse Wonder","score":85.6,"maker":"Meta"},"model_count":5,"drop_count":1,"first_date":"2026-09-01","last_date":"2026-10-12","frontier_gain":6.6,"frontier":[{"date":"2026-09-01","model":"Fable 5.1","score":79,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":81.4,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":81.5,"maker":"anthropic"},{"date":"2026-10-12","model":"Muse Wonder","score":85.6,"maker":"meta"}],"models":[{"model":"Muse Wonder","score":85.6,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":85.6,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":1},{"model":"Muse Wonder Full","score":85.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":2},{"model":"Claude Opus 5.5","score":81.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":3},{"model":"GPT-6 Astra","score":81.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":4},{"model":"Fable 5.1","score":79,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":5}]},{"key":"mmmu-pro-vision","name":"MMMU-Pro Vision","about":null,"type":{"key":"multimodal","label":"Multimodal"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#mmmu-pro-vision","api_url":"https://handyai.news/api/benchmarks/mmmu-pro-vision","image_url":"https://handyai.news/api/benchmarks/mmmu-pro-vision.png","leader":{"model":"GPT-6 Astra","score":86.7,"maker":"OpenAI"},"model_count":5,"drop_count":1,"first_date":"2026-06-26","last_date":"2026-10-26","frontier_gain":5,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Terra","score":81.7,"maker":"openai"},{"date":"2026-09-03","model":"GPT-6 Astra","score":86.7,"maker":"openai"}],"models":[{"model":"GPT-6 Astra","score":86.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":1},{"model":"GPT-6.1 Sol","score":86.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":2},{"model":"GPT-6.1 Terra","score":85.5,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":85.5,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":3},{"model":"GPT-5.6 Terra","score":81.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":4},{"model":"GPT-6 Luna","score":79.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-6.1 Terra","slug":"gpt-61-terra","url":"https://handyai.news/modeldrop/gpt-61-terra","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":5}]},{"key":"personal-superintelligence-index","name":"Personal Superintelligence Index","about":null,"type":{"key":"overall","label":"Overall index","folded":true},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#personal-superintelligence-index","api_url":"https://handyai.news/api/benchmarks/personal-superintelligence-index","image_url":"https://handyai.news/api/benchmarks/personal-superintelligence-index.png","leader":{"model":"Muse Wonder Full","score":59.1,"maker":"Meta"},"model_count":5,"drop_count":1,"first_date":"2026-09-01","last_date":"2026-10-12","frontier_gain":9.2,"frontier":[{"date":"2026-09-01","model":"Fable 5.1","score":49.9,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":54.5,"maker":"openai"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":55.2,"maker":"anthropic"},{"date":"2026-10-12","model":"Muse Wonder Full","score":59.1,"maker":"meta"}],"models":[{"model":"Muse Wonder Full","score":59.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":1},{"model":"Claude Opus 5.5","score":55.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2},{"model":"GPT-6 Astra","score":54.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":3},{"model":"Muse Wonder","score":54.3,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":54.3,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":4},{"model":"Fable 5.1","score":49.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Wonder","slug":"muse-wonder","url":"https://handyai.news/modeldrop/muse-wonder","published":"2026-10-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":5}]},{"key":"swe-sweep","name":"SWE-sweep","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#swe-sweep","api_url":"https://handyai.news/api/benchmarks/swe-sweep","image_url":"https://handyai.news/api/benchmarks/swe-sweep.png","leader":{"model":"Gemini 4 Carbon","score":9.6,"maker":"Google DeepMind"},"model_count":5,"drop_count":1,"first_date":"2026-09-01","last_date":"2026-10-28","frontier_gain":6.7,"frontier":[{"date":"2026-09-01","model":"Fable 5.1","score":2.9,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":9.5,"maker":"openai"},{"date":"2026-10-28","model":"Gemini 4 Carbon","score":9.6,"maker":"google"}],"models":[{"model":"Gemini 4 Carbon","score":9.6,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":9.6,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":1},{"model":"GPT-6 Astra","score":9.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":2},{"model":"Gemini 4 Argon","score":4.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":3},{"model":"Claude Opus 5.5","score":3.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":4},{"model":"Fable 5.1","score":2.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":5}]},{"key":"vals-finance-agent-v2","name":"Vals Finance Agent v2","about":null,"type":{"key":"work","label":"Work"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#vals-finance-agent-v2","api_url":"https://handyai.news/api/benchmarks/vals-finance-agent-v2","image_url":"https://handyai.news/api/benchmarks/vals-finance-agent-v2.png","leader":{"model":"Gemini 3.8 Flash","score":61.4,"maker":"Google DeepMind"},"model_count":5,"drop_count":1,"first_date":"2026-06-26","last_date":"2026-09-02","frontier_gain":7,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Terra","score":54.4,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":58.6,"maker":"anthropic"},{"date":"2026-09-02","model":"Gemini 3.8 Flash","score":61.4,"maker":"google"}],"models":[{"model":"Gemini 3.8 Flash","score":61.4,"own_drop":true,"independent":true,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":61.4,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":1},{"model":"Claude Opus 5","score":58.6,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":2},{"model":"GPT-5.6 Terra","score":54.4,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":3},{"model":"Claude Sonnet 5","score":53.9,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":4},{"model":"GPT-5.6 Sol","score":53.8,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Gemini 3.8 Flash","slug":"gemini-38-flash","url":"https://handyai.news/modeldrop/gemini-38-flash","published":"2026-09-02"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":5}]},{"key":"bench","name":"τ²-Bench","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#bench","api_url":"https://handyai.news/api/benchmarks/bench","image_url":"https://handyai.news/api/benchmarks/bench.png","leader":{"model":"Claude Sonnet 4.5","score":84.7,"maker":"Anthropic"},"model_count":5,"drop_count":1,"first_date":"2025-08-07","last_date":"2025-10-27","frontier_gain":4.6,"frontier":[{"date":"2025-08-07","model":"GPT-5","score":80.1,"maker":"openai"},{"date":"2025-09-29","model":"Claude Sonnet 4.5","score":84.7,"maker":"anthropic"}],"models":[{"model":"Claude Sonnet 4.5","score":84.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":1},{"model":"GPT-5","score":80.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":2},{"model":"MiniMax-M2","score":77.2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":77.2,"reports":1,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m2","rank":3},{"model":"GLM-4.6","score":75.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2025-10-27","dated_by":"first_cited","drop_url":null,"rank":4},{"model":"Kimi K2 0905","score":70.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MiniMax-M2","slug":"minimax-m2","url":"https://handyai.news/modeldrop/minimax-m2","published":"2025-10-27"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2025-09-24","dated_by":"first_cited","drop_url":null,"rank":5}]},{"key":"agents-last-exam-v1-1","name":"Agents' Last Exam v1.1","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#agents-last-exam-v1-1","api_url":"https://handyai.news/api/benchmarks/agents-last-exam-v1-1","image_url":"https://handyai.news/api/benchmarks/agents-last-exam-v1-1.png","leader":{"model":"Gemini 4 Carbon","score":77.4,"maker":"Google DeepMind"},"model_count":4,"drop_count":1,"first_date":"2026-09-03","last_date":"2026-10-28","frontier_gain":4.3,"frontier":[{"date":"2026-09-03","model":"GPT-6 Astra","score":73.1,"maker":"openai"},{"date":"2026-10-28","model":"Gemini 4 Carbon","score":77.4,"maker":"google"}],"models":[{"model":"Gemini 4 Carbon","score":77.4,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":77.4,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":1},{"model":"GPT-6 Astra","score":73.1,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":2},{"model":"Claude Opus 5.5","score":71.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":3},{"model":"Gemini 4 Argon","score":69,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":4}]},{"key":"arenahard","name":"ArenaHard","about":null,"type":{"key":"overall","label":"Overall index","folded":true},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#arenahard","api_url":"https://handyai.news/api/benchmarks/arenahard","image_url":"https://handyai.news/api/benchmarks/arenahard.png","leader":{"model":"Mistral Medium 3","score":97.1,"maker":"Mistral AI"},"model_count":4,"drop_count":1,"first_date":"2024-12-05","last_date":"2025-05-07","frontier_gain":1.7,"frontier":[{"date":"2024-12-05","model":"GPT-4o","score":95.4,"maker":"openai"},{"date":"2025-05-07","model":"Mistral Medium 3","score":97.1,"maker":"mistral"}],"models":[{"model":"Mistral Medium 3","score":97.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":97.1,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2025-05-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-medium-3","rank":1},{"model":"GPT-4o","score":95.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":2},{"model":"Claude 3.7 Sonnet","score":93.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":3},{"model":"Llama 4 Maverick","score":91.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":4}]},{"key":"chartography","name":"Chartography","about":null,"type":{"key":"multimodal","label":"Multimodal"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#chartography","api_url":"https://handyai.news/api/benchmarks/chartography","image_url":"https://handyai.news/api/benchmarks/chartography.png","leader":{"model":"Claude Sonnet 5.5","score":61.6,"maker":"Anthropic"},"model_count":4,"drop_count":1,"first_date":"2025-10-15","last_date":"2026-10-07","frontier_gain":55.2,"frontier":[{"date":"2025-10-15","model":"Claude Haiku 4.5","score":6.4,"maker":"anthropic"},{"date":"2026-09-22","model":"GPT-6 Luna","score":29.1,"maker":"openai"},{"date":"2026-09-28","model":"Claude Sonnet 5.5","score":61.6,"maker":"anthropic"}],"models":[{"model":"Claude Sonnet 5.5","score":61.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":1},{"model":"Claude Haiku 5.5","score":46.4,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":46.4,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":2},{"model":"GPT-6 Luna","score":29.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":3},{"model":"Claude Haiku 4.5","score":6.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Claude Haiku 5.5","slug":"claude-haiku-55","url":"https://handyai.news/modeldrop/claude-haiku-55","published":"2026-10-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":4}]},{"key":"chartqa","name":"ChartQA","about":null,"type":{"key":"multimodal","label":"Multimodal"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#chartqa","api_url":"https://handyai.news/api/benchmarks/chartqa","image_url":"https://handyai.news/api/benchmarks/chartqa.png","leader":{"model":"Llama 4 Maverick","score":90,"maker":"Meta"},"model_count":4,"drop_count":1,"first_date":"2024-12-05","last_date":"2025-04-05","frontier_gain":4.3,"frontier":[{"date":"2024-12-05","model":"GPT-4o","score":85.7,"maker":"openai"},{"date":"2024-12-11","model":"Gemini 2.0 Flash","score":88.3,"maker":"google"},{"date":"2025-04-05","model":"Llama 4 Maverick","score":90,"maker":"meta"}],"models":[{"model":"Llama 4 Maverick","score":90,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":90,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":1},{"model":"Llama 4 Scout","score":88.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":88.8,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":2},{"model":"Gemini 2.0 Flash","score":88.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":3},{"model":"GPT-4o","score":85.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":4}]},{"key":"claw-eval-pass-3","name":"Claw-Eval pass^3","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#claw-eval-pass-3","api_url":"https://handyai.news/api/benchmarks/claw-eval-pass-3","image_url":"https://handyai.news/api/benchmarks/claw-eval-pass-3.png","leader":{"model":"Nanbeige4.3-33B","score":78.3,"maker":"Nanbeige"},"model_count":4,"drop_count":1,"first_date":"2026-04-02","last_date":"2026-10-24","frontier_gain":34.1,"frontier":[{"date":"2026-04-02","model":"Gemma 4","score":44.2,"maker":"google"},{"date":"2026-08-12","model":"Qwen3.8","score":75.2,"maker":"qwen"},{"date":"2026-10-24","model":"Nanbeige4.3-33B","score":78.3,"maker":"nanbeige"}],"models":[{"model":"Nanbeige4.3-33B","score":78.3,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":78.3,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":1},{"model":"Qwen3.8","score":75.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":2},{"model":"Muse Glimmer-30B","score":57.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":3},{"model":"Gemma 4","score":44.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":4}]},{"key":"cursorbench-3-2","name":"CursorBench 3.2","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#cursorbench-3-2","api_url":"https://handyai.news/api/benchmarks/cursorbench-3-2","image_url":"https://handyai.news/api/benchmarks/cursorbench-3-2.png","leader":{"model":"Fable 5","score":70.5,"maker":"Anthropic"},"model_count":4,"drop_count":1,"first_date":"2026-06-09","last_date":"2026-08-12","frontier_gain":0,"frontier":[{"date":"2026-06-09","model":"Fable 5","score":70.5,"maker":"anthropic"}],"models":[{"model":"Fable 5","score":70.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":1},{"model":"Grok 4.6","score":69.9,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":69.9,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":2},{"model":"GPT-5.6 Sol","score":67.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":3},{"model":"Grok 4.5","score":66.7,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4.6","slug":"grok-46","url":"https://handyai.news/modeldrop/grok-46","published":"2026-08-12"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":4}]},{"key":"deepplanning","name":"DeepPlanning","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#deepplanning","api_url":"https://handyai.news/api/benchmarks/deepplanning","image_url":"https://handyai.news/api/benchmarks/deepplanning.png","leader":{"model":"Nanbeige4.3-33B","score":63.3,"maker":"Nanbeige"},"model_count":4,"drop_count":1,"first_date":"2026-04-02","last_date":"2026-10-24","frontier_gain":43.9,"frontier":[{"date":"2026-04-02","model":"Gemma 4","score":19.4,"maker":"google"},{"date":"2026-08-12","model":"Qwen3.8","score":49.2,"maker":"qwen"},{"date":"2026-10-24","model":"Nanbeige4.3-33B","score":63.3,"maker":"nanbeige"}],"models":[{"model":"Nanbeige4.3-33B","score":63.3,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":63.3,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":1},{"model":"Qwen3.8","score":49.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":2},{"model":"Muse Glimmer-30B","score":21.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":3},{"model":"Gemma 4","score":19.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":4}]},{"key":"docvqa","name":"DocVQA","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#docvqa","api_url":"https://handyai.news/api/benchmarks/docvqa","image_url":"https://handyai.news/api/benchmarks/docvqa.png","leader":{"model":"Mistral Medium 3","score":95.3,"maker":"Mistral AI"},"model_count":4,"drop_count":1,"first_date":"2024-12-05","last_date":"2025-05-07","frontier_gain":9.4,"frontier":[{"date":"2024-12-05","model":"GPT-4o","score":85.9,"maker":"openai"},{"date":"2025-04-05","model":"Llama 4 Maverick","score":94.1,"maker":"meta"},{"date":"2025-05-07","model":"Mistral Medium 3","score":95.3,"maker":"mistral"}],"models":[{"model":"Mistral Medium 3","score":95.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":95.3,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2025-05-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-medium-3","rank":1},{"model":"Llama 4 Maverick","score":94.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":2},{"model":"GPT-4o","score":85.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":3},{"model":"Claude 3.7 Sonnet","score":84.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":4}]},{"key":"eebench","name":"EEBench","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#eebench","api_url":"https://handyai.news/api/benchmarks/eebench","image_url":"https://handyai.news/api/benchmarks/eebench.png","leader":{"model":"Grok 4.7","score":64,"maker":"SpaceXAI"},"model_count":4,"drop_count":1,"first_date":"2026-06-26","last_date":"2026-09-21","frontier_gain":24.6,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":39.4,"maker":"openai"},{"date":"2026-08-12","model":"Grok 4.6","score":53,"maker":"xai"},{"date":"2026-09-01","model":"Fable 5.1","score":56.4,"maker":"anthropic"},{"date":"2026-09-21","model":"Grok 4.7","score":64,"maker":"xai"}],"models":[{"model":"Grok 4.7","score":64,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":64,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":1},{"model":"Fable 5.1","score":56.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":2},{"model":"Grok 4.6","score":53,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-46","rank":3},{"model":"GPT-5.6 Sol","score":39.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Grok 4.7","slug":"grok-47","url":"https://handyai.news/modeldrop/grok-47","published":"2026-09-21"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":4}]},{"key":"finance-agent-v2","name":"Finance Agent v2","about":null,"type":{"key":"work","label":"Work"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#finance-agent-v2","api_url":"https://handyai.news/api/benchmarks/finance-agent-v2","image_url":"https://handyai.news/api/benchmarks/finance-agent-v2.png","leader":{"model":"GLM-5.3","score":55.8,"maker":"Z.ai"},"model_count":4,"drop_count":1,"first_date":"2026-07-16","last_date":"2026-10-06","frontier_gain":2.7,"frontier":[{"date":"2026-07-16","model":"Kimi K3","score":53.1,"maker":"moonshot"},{"date":"2026-08-14","model":"GLM-5.3","score":55.8,"maker":"zai"}],"models":[{"model":"GLM-5.3","score":55.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":1},{"model":"Mistral Large 4","score":54.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":54.7,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":2},{"model":"GPT-6 Astra","score":53.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":3},{"model":"Kimi K3","score":53.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":4}]},{"key":"gdm-mrcr-v2","name":"GDM-MRCR v2","about":null,"type":{"key":"long-context","label":"Long context"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#gdm-mrcr-v2","api_url":"https://handyai.news/api/benchmarks/gdm-mrcr-v2","image_url":"https://handyai.news/api/benchmarks/gdm-mrcr-v2.png","leader":{"model":"Gemini 3.6 Flash","score":91.8,"maker":"Google DeepMind"},"model_count":4,"drop_count":1,"first_date":"2026-06-26","last_date":"2026-07-21","frontier_gain":17,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Luna","score":74.8,"maker":"openai"},{"date":"2026-07-09","model":"Grok 4.5","score":81.4,"maker":"xai"},{"date":"2026-07-21","model":"Gemini 3.6 Flash","score":91.8,"maker":"google"}],"models":[{"model":"Gemini 3.6 Flash","score":91.8,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":91.8,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":1},{"model":"Grok 4.5","score":81.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":2},{"model":"GPT-5.6 Luna","score":74.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":3},{"model":"Claude Sonnet 5","score":71.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":4}]},{"key":"gdpval-rubrics","name":"GDPval rubrics","about":null,"type":{"key":"work","label":"Work"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#gdpval-rubrics","api_url":"https://handyai.news/api/benchmarks/gdpval-rubrics","image_url":"https://handyai.news/api/benchmarks/gdpval-rubrics.png","leader":{"model":"Qwen3.8","score":86.4,"maker":"Alibaba"},"model_count":4,"drop_count":1,"first_date":"2026-04-02","last_date":"2026-10-24","frontier_gain":15,"frontier":[{"date":"2026-04-02","model":"Gemma 4","score":71.4,"maker":"google"},{"date":"2026-08-12","model":"Qwen3.8","score":86.4,"maker":"qwen"}],"models":[{"model":"Qwen3.8","score":86.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":1},{"model":"Nanbeige4.3-33B","score":84.8,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":84.8,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":2},{"model":"Muse Glimmer-30B","score":75.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":3},{"model":"Gemma 4","score":71.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":4}]},{"key":"healthbench-hard","name":"HealthBench Hard","about":null,"type":{"key":"work","label":"Work"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#healthbench-hard","api_url":"https://handyai.news/api/benchmarks/healthbench-hard","image_url":"https://handyai.news/api/benchmarks/healthbench-hard.png","leader":{"model":"Muse Spark","score":42.8,"maker":"Meta"},"model_count":4,"drop_count":1,"first_date":"2026-02-05","last_date":"2026-04-08","frontier_gain":28,"frontier":[{"date":"2026-02-05","model":"Claude Opus 4.6","score":14.8,"maker":"anthropic"},{"date":"2026-02-19","model":"Gemini 3.1 Pro","score":20.6,"maker":"google"},{"date":"2026-03-05","model":"GPT-5.4","score":40.1,"maker":"openai"},{"date":"2026-04-08","model":"Muse Spark","score":42.8,"maker":"meta"}],"models":[{"model":"Muse Spark","score":42.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":42.8,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-04-08","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark","rank":1},{"model":"GPT-5.4","score":40.1,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-03-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-54","rank":2},{"model":"Gemini 3.1 Pro","score":20.6,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-02-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-31-pro","rank":3},{"model":"Claude Opus 4.6","score":14.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark","slug":"muse-spark","url":"https://handyai.news/modeldrop/muse-spark","published":"2026-04-08"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":4}]},{"key":"hle-diamond-no-tools","name":"HLE-Diamond (no tools)","about":null,"type":{"key":"reasoning","label":"Reasoning"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#hle-diamond-no-tools","api_url":"https://handyai.news/api/benchmarks/hle-diamond-no-tools","image_url":"https://handyai.news/api/benchmarks/hle-diamond-no-tools.png","leader":{"model":"Muse Spark Open","score":26.1,"maker":"Meta"},"model_count":4,"drop_count":1,"first_date":"2026-07-16","last_date":"2026-10-15","frontier_gain":3.9,"frontier":[{"date":"2026-07-16","model":"Kimi K3","score":22.2,"maker":"moonshot"},{"date":"2026-10-15","model":"Muse Spark Open","score":26.1,"maker":"meta"}],"models":[{"model":"Muse Spark Open","score":26.1,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":26.1,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":1},{"model":"Kimi K3","score":22.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":2},{"model":"Qwen3.8-2.4T-A95B","score":17.3,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"first_cited","drop_url":null,"rank":3},{"model":"GLM-5.3","score":16.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark Open","slug":"muse-spark-open","url":"https://handyai.news/modeldrop/muse-spark-open","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":4}]},{"key":"humaneval","name":"HumanEval","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#humaneval","api_url":"https://handyai.news/api/benchmarks/humaneval","image_url":"https://handyai.news/api/benchmarks/humaneval.png","leader":{"model":"Mistral Medium 3","score":92.1,"maker":"Mistral AI"},"model_count":4,"drop_count":1,"first_date":"2024-12-05","last_date":"2025-05-07","frontier_gain":0.6,"frontier":[{"date":"2024-12-05","model":"GPT-4o","score":91.5,"maker":"openai"},{"date":"2025-02-24","model":"Claude 3.7 Sonnet","score":92.1,"maker":"anthropic"}],"models":[{"model":"Mistral Medium 3","score":92.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":92.1,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2025-05-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-medium-3","rank":1},{"model":"Claude 3.7 Sonnet","score":92.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-02-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-37-sonnet","rank":2},{"model":"GPT-4o","score":91.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":3},{"model":"Llama 4 Maverick","score":85.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Medium 3","slug":"mistral-medium-3","url":"https://handyai.news/modeldrop/mistral-medium-3","published":"2025-05-07"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":4}]},{"key":"matharena-apex","name":"MathArena Apex","about":null,"type":{"key":"math","label":"Math"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#matharena-apex","api_url":"https://handyai.news/api/benchmarks/matharena-apex","image_url":"https://handyai.news/api/benchmarks/matharena-apex.png","leader":{"model":"Gemini 3 Pro","score":23.4,"maker":"Google DeepMind"},"model_count":4,"drop_count":1,"first_date":"2025-03-25","last_date":"2025-11-18","frontier_gain":22.9,"frontier":[{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":0.5,"maker":"google"},{"date":"2025-09-29","model":"Claude Sonnet 4.5","score":1.6,"maker":"anthropic"},{"date":"2025-11-18","model":"Gemini 3 Pro","score":23.4,"maker":"google"}],"models":[{"model":"Gemini 3 Pro","score":23.4,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":23.4,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-11-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-3-pro","rank":1},{"model":"Claude Sonnet 4.5","score":1.6,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":2},{"model":"GPT-5.1","score":1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":3},{"model":"Gemini 2.5 Pro","score":0.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 3 Pro","slug":"gemini-3-pro","url":"https://handyai.news/modeldrop/gemini-3-pro","published":"2025-11-18"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":4}]},{"key":"mathvista","name":"MathVista","about":null,"type":{"key":"multimodal","label":"Multimodal"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#mathvista","api_url":"https://handyai.news/api/benchmarks/mathvista","image_url":"https://handyai.news/api/benchmarks/mathvista.png","leader":{"model":"Llama 4 Maverick","score":73.7,"maker":"Meta"},"model_count":4,"drop_count":1,"first_date":"2024-12-05","last_date":"2025-04-05","frontier_gain":9.9,"frontier":[{"date":"2024-12-05","model":"GPT-4o","score":63.8,"maker":"openai"},{"date":"2024-12-11","model":"Gemini 2.0 Flash","score":73.1,"maker":"google"},{"date":"2025-04-05","model":"Llama 4 Maverick","score":73.7,"maker":"meta"}],"models":[{"model":"Llama 4 Maverick","score":73.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":73.7,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":1},{"model":"Gemini 2.0 Flash","score":73.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":2},{"model":"Llama 4 Scout","score":70.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":70.7,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2025-04-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-4","rank":3},{"model":"GPT-4o","score":63.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Llama 4 Maverick & Scout","slug":"llama-4","url":"https://handyai.news/modeldrop/llama-4","published":"2025-04-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":4}]},{"key":"meta-internal-coding-bench","name":"Meta Internal Coding Bench","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#meta-internal-coding-bench","api_url":"https://handyai.news/api/benchmarks/meta-internal-coding-bench","image_url":"https://handyai.news/api/benchmarks/meta-internal-coding-bench.png","leader":{"model":"Claude Opus 5","score":79.4,"maker":"Anthropic"},"model_count":4,"drop_count":1,"first_date":"2026-06-26","last_date":"2026-08-05","frontier_gain":14,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Terra","score":65.4,"maker":"openai"},{"date":"2026-07-09","model":"Muse Spark 1.1","score":68.3,"maker":"meta"},{"date":"2026-07-24","model":"Claude Opus 5","score":79.4,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5","score":79.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark 1.2","slug":"muse-spark-12","url":"https://handyai.news/modeldrop/muse-spark-12","published":"2026-08-05"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":1},{"model":"Muse Spark 1.2","score":70.6,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark 1.2","slug":"muse-spark-12","url":"https://handyai.news/modeldrop/muse-spark-12","published":"2026-08-05"},"launch_score":70.6,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-12","rank":2},{"model":"Muse Spark 1.1","score":68.3,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark 1.2","slug":"muse-spark-12","url":"https://handyai.news/modeldrop/muse-spark-12","published":"2026-08-05"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-11","rank":3},{"model":"GPT-5.6 Terra","score":65.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark 1.2","slug":"muse-spark-12","url":"https://handyai.news/modeldrop/muse-spark-12","published":"2026-08-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":4}]},{"key":"mle-bench","name":"MLE-Bench","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#mle-bench","api_url":"https://handyai.news/api/benchmarks/mle-bench","image_url":"https://handyai.news/api/benchmarks/mle-bench.png","leader":{"model":"Claude Sonnet 5","score":66.9,"maker":"Anthropic"},"model_count":4,"drop_count":1,"first_date":"2026-06-26","last_date":"2026-07-21","frontier_gain":19.3,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Luna","score":47.6,"maker":"openai"},{"date":"2026-06-30","model":"Claude Sonnet 5","score":66.9,"maker":"anthropic"}],"models":[{"model":"Claude Sonnet 5","score":66.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-30","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-5","rank":1},{"model":"Gemini 3.6 Flash","score":63.9,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":63.9,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-07-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-36-flash","rank":2},{"model":"GPT-5.6 Luna","score":47.6,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":3},{"model":"Grok 4.5","score":43.2,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Gemini 3.6 Flash","slug":"gemini-36-flash","url":"https://handyai.news/modeldrop/gemini-36-flash","published":"2026-07-21"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-45","rank":4}]},{"key":"paperbench","name":"PaperBench","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#paperbench","api_url":"https://handyai.news/api/benchmarks/paperbench","image_url":"https://handyai.news/api/benchmarks/paperbench.png","leader":{"model":"Qwen3.8-Max","score":93,"maker":"Alibaba"},"model_count":4,"drop_count":1,"first_date":"2026-05-28","last_date":"2026-07-19","frontier_gain":12.7,"frontier":[{"date":"2026-05-28","model":"Claude Opus 4.8","score":80.3,"maker":"anthropic"},{"date":"2026-06-09","model":"Fable 5","score":88.8,"maker":"anthropic"},{"date":"2026-06-26","model":"GPT-5.6 Sol","score":90.5,"maker":"openai"},{"date":"2026-07-19","model":"Qwen3.8-Max","score":93,"maker":"qwen"}],"models":[{"model":"Qwen3.8-Max","score":93,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":93,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":1},{"model":"GPT-5.6 Sol","score":90.5,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":2},{"model":"Fable 5","score":88.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":3},{"model":"Claude Opus 4.8","score":80.3,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Qwen3.8-Max","slug":"qwen38-max","url":"https://handyai.news/modeldrop/qwen38-max","published":"2026-07-19"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-48","rank":4}]},{"key":"sec-bench-pro","name":"SEC-Bench Pro","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#sec-bench-pro","api_url":"https://handyai.news/api/benchmarks/sec-bench-pro","image_url":"https://handyai.news/api/benchmarks/sec-bench-pro.png","leader":{"model":"GPT-5.6 Sol","score":71.2,"maker":"OpenAI"},"model_count":4,"drop_count":1,"first_date":"2026-04-23","last_date":"2026-06-26","frontier_gain":25.4,"frontier":[{"date":"2026-04-23","model":"GPT-5.5","score":45.8,"maker":"openai"},{"date":"2026-06-26","model":"GPT-5.6 Sol","score":71.2,"maker":"openai"}],"models":[{"model":"GPT-5.6 Sol","score":71.2,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"GPT-5.6 Sol, Terra, & Luna","slug":"gpt-56-sol-terra-and-luna","url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","published":"2026-06-26"},"launch_score":71.2,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":1},{"model":"GPT-5.6 Terra","score":57.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-5.6 Sol, Terra, & Luna","slug":"gpt-56-sol-terra-and-luna","url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","published":"2026-06-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":2},{"model":"GPT-5.6 Luna","score":48.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"GPT-5.6 Sol, Terra, & Luna","slug":"gpt-56-sol-terra-and-luna","url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","published":"2026-06-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":3},{"model":"GPT-5.5","score":45.8,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.6 Sol, Terra, & Luna","slug":"gpt-56-sol-terra-and-luna","url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","published":"2026-06-26"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-04-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-55","rank":4}]},{"key":"simpleqa-hallucination-rate","name":"SimpleQA hallucination rate","about":null,"type":{"key":"knowledge","label":"Knowledge"},"unit":"%","lower_is_better":true,"url":"https://handyai.news/benchmarks#simpleqa-hallucination-rate","api_url":"https://handyai.news/api/benchmarks/simpleqa-hallucination-rate","image_url":"https://handyai.news/api/benchmarks/simpleqa-hallucination-rate.png","leader":{"model":"GPT-4.5","score":37.1,"maker":"OpenAI"},"model_count":4,"drop_count":1,"first_date":"2024-12-05","last_date":"2025-02-27","frontier_gain":6.9,"frontier":[{"date":"2024-12-05","model":"o1","score":44,"maker":"openai"},{"date":"2025-02-27","model":"GPT-4.5","score":37.1,"maker":"openai"}],"models":[{"model":"GPT-4.5","score":37.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":37.1,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":1},{"model":"o1","score":44,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":2},{"model":"GPT-4o","score":61.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":3},{"model":"o3-mini","score":80.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":4}]},{"key":"skillsbench","name":"SkillsBench","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#skillsbench","api_url":"https://handyai.news/api/benchmarks/skillsbench","image_url":"https://handyai.news/api/benchmarks/skillsbench.png","leader":{"model":"Nanbeige4.3-33B","score":61.1,"maker":"Nanbeige"},"model_count":4,"drop_count":1,"first_date":"2026-04-02","last_date":"2026-10-24","frontier_gain":32.2,"frontier":[{"date":"2026-04-02","model":"Gemma 4","score":28.9,"maker":"google"},{"date":"2026-08-12","model":"Qwen3.8","score":56.4,"maker":"qwen"},{"date":"2026-10-24","model":"Nanbeige4.3-33B","score":61.1,"maker":"nanbeige"}],"models":[{"model":"Nanbeige4.3-33B","score":61.1,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":61.1,"reports":1,"maker":{"key":"nanbeige","name":"Nanbeige","logo_url":"https://handyai.news/ui/makers/nanbeige.svg"},"open_weights":true,"released":"2026-10-24","dated_by":"own_drop","drop_url":null,"rank":1},{"model":"Qwen3.8","score":56.4,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":2},{"model":"Muse Glimmer-30B","score":39.7,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-12","dated_by":"first_cited","drop_url":null,"rank":3},{"model":"Gemma 4","score":28.9,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Nanbeige4.3-33B model card","slug":"nanbeige43-33b","url":"https://huggingface.co/Nanbeige/Nanbeige4.3-33B","published":"2026-10-24"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":4}]},{"key":"supergpqa","name":"SuperGPQA","about":null,"type":{"key":"reasoning","label":"Reasoning"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#supergpqa","api_url":"https://handyai.news/api/benchmarks/supergpqa","image_url":"https://handyai.news/api/benchmarks/supergpqa.png","leader":{"model":"Qwen3-Max","score":65.1,"maker":"Alibaba"},"model_count":4,"drop_count":1,"first_date":"2025-05-22","last_date":"2025-09-24","frontier_gain":8.6,"frontier":[{"date":"2025-05-22","model":"Claude Opus 4","score":56.5,"maker":"anthropic"},{"date":"2025-08-21","model":"DeepSeek V3.1","score":59.8,"maker":"deepseek"},{"date":"2025-09-24","model":"Qwen3-Max","score":65.1,"maker":"qwen"}],"models":[{"model":"Qwen3-Max","score":65.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":65.1,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2025-09-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3-max","rank":1},{"model":"Qwen3-235B-A22B-Instruct-2507","score":62.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-09-24","dated_by":"first_cited","drop_url":null,"rank":2},{"model":"DeepSeek V3.1","score":59.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-08-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v31","rank":3},{"model":"Claude Opus 4","score":56.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":4}]},{"key":"usamo-2025","name":"USAMO 2025","about":null,"type":{"key":"math","label":"Math"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#usamo-2025","api_url":"https://handyai.news/api/benchmarks/usamo-2025","image_url":"https://handyai.news/api/benchmarks/usamo-2025.png","leader":{"model":"Gemini Deep Think","score":49.4,"maker":"Google DeepMind"},"model_count":4,"drop_count":1,"first_date":"2025-03-25","last_date":"2025-07-09","frontier_gain":14.9,"frontier":[{"date":"2025-03-25","model":"Gemini 2.5 Pro","score":34.5,"maker":"google"},{"date":"2025-07-09","model":"Gemini Deep Think","score":49.4,"maker":"google"}],"models":[{"model":"Gemini Deep Think","score":49.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4","slug":"grok-4","url":"https://handyai.news/modeldrop/grok-4","published":"2025-07-09"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"first_cited","drop_url":null,"rank":1},{"model":"Grok 4","score":37.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4","slug":"grok-4","url":"https://handyai.news/modeldrop/grok-4","published":"2025-07-09"},"launch_score":37.5,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":2},{"model":"Gemini 2.5 Pro","score":34.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4","slug":"grok-4","url":"https://handyai.news/modeldrop/grok-4","published":"2025-07-09"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2025-03-25","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-25-pro","rank":3},{"model":"o3","score":21.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4","slug":"grok-4","url":"https://handyai.news/modeldrop/grok-4","published":"2025-07-09"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":4}]},{"key":"bench-weighted","name":"τ²-Bench (Weighted)","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#bench-weighted","api_url":"https://handyai.news/api/benchmarks/bench-weighted","image_url":"https://handyai.news/api/benchmarks/bench-weighted.png","leader":{"model":"Qwen3-Max","score":74.8,"maker":"Alibaba"},"model_count":4,"drop_count":1,"first_date":"2025-05-22","last_date":"2025-09-24","frontier_gain":7.1,"frontier":[{"date":"2025-05-22","model":"Claude Opus 4","score":67.7,"maker":"anthropic"},{"date":"2025-09-24","model":"Qwen3-Max","score":74.8,"maker":"qwen"}],"models":[{"model":"Qwen3-Max","score":74.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":74.8,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2025-09-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen3-max","rank":1},{"model":"Claude Opus 4","score":67.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-05-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-4","rank":2},{"model":"Qwen3-235B-A22B-Instruct-2507","score":52.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-09-24","dated_by":"first_cited","drop_url":null,"rank":3},{"model":"DeepSeek V3.1","score":46.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Qwen3-Max","slug":"qwen3-max","url":"https://handyai.news/modeldrop/qwen3-max","published":"2025-09-24"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-08-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v31","rank":4}]},{"key":"aime-2024-4-4-reliability","name":"AIME 2024 4/4 reliability","about":null,"type":{"key":"math","label":"Math"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#aime-2024-4-4-reliability","api_url":"https://handyai.news/api/benchmarks/aime-2024-4-4-reliability","image_url":"https://handyai.news/api/benchmarks/aime-2024-4-4-reliability.png","leader":{"model":"o3-pro","score":90,"maker":"OpenAI"},"model_count":3,"drop_count":1,"first_date":"2025-04-16","last_date":"2025-06-10","frontier_gain":10,"frontier":[{"date":"2025-04-16","model":"o3","score":80,"maker":"openai"},{"date":"2025-06-10","model":"o3-pro","score":90,"maker":"openai"}],"models":[{"model":"o3-pro","score":90,"own_drop":true,"independent":false,"setting":"","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":90,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-06-10","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-pro","rank":1},{"model":"o3","score":80,"own_drop":false,"independent":false,"setting":"medium","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":2},{"model":"o1-pro","score":80,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"first_cited","drop_url":null,"rank":3}]},{"key":"arc-agi-3","name":"ARC-AGI-3","about":null,"type":{"key":"reasoning","label":"Reasoning"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#arc-agi-3","api_url":"https://handyai.news/api/benchmarks/arc-agi-3","image_url":"https://handyai.news/api/benchmarks/arc-agi-3.png","leader":{"model":"GPT-6 Astra","score":99.9,"maker":"OpenAI"},"model_count":3,"drop_count":1,"first_date":"2026-06-26","last_date":"2026-09-03","frontier_gain":92.1,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":7.8,"maker":"openai"},{"date":"2026-07-24","model":"Claude Opus 5","score":30.2,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":99.9,"maker":"openai"}],"models":[{"model":"GPT-6 Astra","score":99.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-6 Astra","slug":"gpt-6-astra","url":"https://handyai.news/modeldrop/gpt-6-astra","published":"2026-09-03"},"launch_score":99.9,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":1},{"model":"Claude Opus 5","score":30.2,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-6 Astra","slug":"gpt-6-astra","url":"https://handyai.news/modeldrop/gpt-6-astra","published":"2026-09-03"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-07-24","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-5","rank":2},{"model":"GPT-5.6 Sol","score":7.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-6 Astra","slug":"gpt-6-astra","url":"https://handyai.news/modeldrop/gpt-6-astra","published":"2026-09-03"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":3}]},{"key":"arena-ai-text-score","name":"Arena AI text score","about":null,"type":{"key":"overall","label":"Overall index","folded":true},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#arena-ai-text-score","api_url":"https://handyai.news/api/benchmarks/arena-ai-text-score","image_url":"https://handyai.news/api/benchmarks/arena-ai-text-score.png","leader":{"model":"Gemma 4","score":1452,"maker":"Google DeepMind"},"model_count":3,"drop_count":1,"first_date":"2026-04-02","last_date":"2026-04-02","frontier_gain":0,"frontier":[{"date":"2026-04-02","model":"Gemma 4","score":1452,"maker":"google"}],"models":[{"model":"Gemma 4","score":1452,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":1452,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":1},{"model":"Gemma 4 26B A4B","score":1441,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemma-4","rank":2},{"model":"Gemma 3 27B","score":1365,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemma 4","slug":"gemma-4","url":"https://handyai.news/modeldrop/gemma-4","published":"2026-04-02"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":true,"released":"2026-04-02","dated_by":"first_cited","drop_url":null,"rank":3}]},{"key":"arena-hard","name":"Arena Hard","about":null,"type":{"key":"overall","label":"Overall index","folded":true},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#arena-hard","api_url":"https://handyai.news/api/benchmarks/arena-hard","image_url":"https://handyai.news/api/benchmarks/arena-hard.png","leader":{"model":"DeepSeek R1","score":92,"maker":"DeepSeek"},"model_count":3,"drop_count":1,"first_date":"2024-12-26","last_date":"2025-04-07","frontier_gain":25.8,"frontier":[{"date":"2024-12-26","model":"Llama 3.1 405B","score":66.2,"maker":"meta"},{"date":"2025-01-20","model":"DeepSeek R1","score":92,"maker":"deepseek"}],"models":[{"model":"DeepSeek R1","score":92,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Llama 3.1 Nemotron Ultra","slug":"llama-31-nemotron-ultra","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","published":"2025-04-07"},"launch_score":null,"reports":1,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2025-01-20","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-r1","rank":1},{"model":"Llama 3.1 Nemotron Ultra","score":87,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Llama 3.1 Nemotron Ultra","slug":"llama-31-nemotron-ultra","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","published":"2025-04-07"},"launch_score":87,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","rank":2},{"model":"Llama 3.1 405B","score":66.2,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Llama 3.1 Nemotron Ultra","slug":"llama-31-nemotron-ultra","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","published":"2025-04-07"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"first_cited","drop_url":null,"rank":3}]},{"key":"biomysterybench-hard","name":"BioMysteryBench hard","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#biomysterybench-hard","api_url":"https://handyai.news/api/benchmarks/biomysterybench-hard","image_url":"https://handyai.news/api/benchmarks/biomysterybench-hard.png","leader":{"model":"Mythos 5","score":46.1,"maker":"Anthropic"},"model_count":3,"drop_count":1,"first_date":"2026-04-07","last_date":"2026-06-09","frontier_gain":16.5,"frontier":[{"date":"2026-04-07","model":"Claude Mythos Preview","score":29.6,"maker":"anthropic"},{"date":"2026-05-28","model":"Claude Opus 4.8","score":40,"maker":"anthropic"},{"date":"2026-06-09","model":"Mythos 5","score":46.1,"maker":"anthropic"}],"models":[{"model":"Mythos 5","score":46.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":46.1,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":1},{"model":"Claude Opus 4.8","score":40,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-48","rank":2},{"model":"Claude Mythos Preview","score":29.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":3}]},{"key":"biomysterybench-human-solved","name":"BioMysteryBench human-solved","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#biomysterybench-human-solved","api_url":"https://handyai.news/api/benchmarks/biomysterybench-human-solved","image_url":"https://handyai.news/api/benchmarks/biomysterybench-human-solved.png","leader":{"model":"Mythos 5","score":83.9,"maker":"Anthropic"},"model_count":3,"drop_count":1,"first_date":"2026-04-07","last_date":"2026-06-09","frontier_gain":1.3,"frontier":[{"date":"2026-04-07","model":"Claude Mythos Preview","score":82.6,"maker":"anthropic"},{"date":"2026-06-09","model":"Mythos 5","score":83.9,"maker":"anthropic"}],"models":[{"model":"Mythos 5","score":83.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":83.9,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-06-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-5-mythos-5","rank":1},{"model":"Claude Mythos Preview","score":82.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":2},{"model":"Claude Opus 4.8","score":80.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Fable 5 / Mythos 5","slug":"fable-5-mythos-5","url":"https://handyai.news/modeldrop/fable-5-mythos-5","published":"2026-06-09"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-05-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-48","rank":3}]},{"key":"codeforces-4-4-reliability","name":"Codeforces 4/4 reliability","about":null,"type":{"key":"coding","label":"Coding"},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#codeforces-4-4-reliability","api_url":"https://handyai.news/api/benchmarks/codeforces-4-4-reliability","image_url":"https://handyai.news/api/benchmarks/codeforces-4-4-reliability.png","leader":{"model":"o3-pro","score":2301,"maker":"OpenAI"},"model_count":3,"drop_count":1,"first_date":"2025-04-16","last_date":"2025-06-10","frontier_gain":290,"frontier":[{"date":"2025-04-16","model":"o3","score":2011,"maker":"openai"},{"date":"2025-06-10","model":"o3-pro","score":2301,"maker":"openai"}],"models":[{"model":"o3-pro","score":2301,"own_drop":true,"independent":false,"setting":"","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":2301,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-06-10","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-pro","rank":1},{"model":"o3","score":2011,"own_drop":false,"independent":false,"setting":"medium","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":2},{"model":"o1-pro","score":1423,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"first_cited","drop_url":null,"rank":3}]},{"key":"creative-writing-v3","name":"Creative Writing v3","about":null,"type":{"key":"other","label":"Other"},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#creative-writing-v3","api_url":"https://handyai.news/api/benchmarks/creative-writing-v3","image_url":"https://handyai.news/api/benchmarks/creative-writing-v3.png","leader":{"model":"Grok 4.1","score":1721.9,"maker":"SpaceXAI"},"model_count":3,"drop_count":1,"first_date":"2025-04-16","last_date":"2025-11-17","frontier_gain":25.5,"frontier":[{"date":"2025-04-16","model":"o3","score":1696.4,"maker":"openai"},{"date":"2025-11-17","model":"Grok 4.1","score":1721.9,"maker":"xai"}],"models":[{"model":"Grok 4.1","score":1721.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":1721.9,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-11-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-41","rank":1},{"model":"o3","score":1696.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":2},{"model":"Claude Sonnet 4.5","score":1648.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-45","rank":3}]},{"key":"cve-bench","name":"CVE-Bench","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#cve-bench","api_url":"https://handyai.news/api/benchmarks/cve-bench","image_url":"https://handyai.news/api/benchmarks/cve-bench.png","leader":{"model":"GPT-5.2-Codex","score":87,"maker":"OpenAI"},"model_count":3,"drop_count":1,"first_date":"2025-11-19","last_date":"2025-12-18","frontier_gain":7,"frontier":[{"date":"2025-11-19","model":"GPT-5.1-Codex-Max","score":80,"maker":"openai"},{"date":"2025-12-18","model":"GPT-5.2-Codex","score":87,"maker":"openai"}],"models":[{"model":"GPT-5.2-Codex","score":87,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":87,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52-codex","rank":1},{"model":"GPT-5.1-Codex-Max","score":80,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51-codex-max","rank":2},{"model":"GPT-5.2","score":69,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":3}]},{"key":"cyber-range","name":"Cyber Range","about":null,"type":{"key":"cyber","label":"Cyber"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#cyber-range","api_url":"https://handyai.news/api/benchmarks/cyber-range","image_url":"https://handyai.news/api/benchmarks/cyber-range.png","leader":{"model":"GPT-5.1-Codex-Max","score":81.8,"maker":"OpenAI"},"model_count":3,"drop_count":1,"first_date":"2025-11-19","last_date":"2025-12-18","frontier_gain":0,"frontier":[{"date":"2025-11-19","model":"GPT-5.1-Codex-Max","score":81.8,"maker":"openai"}],"models":[{"model":"GPT-5.1-Codex-Max","score":81.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51-codex-max","rank":1},{"model":"GPT-5.2-Codex","score":72.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":72.7,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52-codex","rank":2},{"model":"GPT-5.2","score":63.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":3}]},{"key":"cybergym-level-1-leaderboard","name":"CyberGym Level 1 leaderboard","about":null,"type":{"key":"cyber","label":"Cyber"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#cybergym-level-1-leaderboard","api_url":"https://handyai.news/api/benchmarks/cybergym-level-1-leaderboard","image_url":"https://handyai.news/api/benchmarks/cybergym-level-1-leaderboard.png","leader":{"model":"Wiz Atlas","score":90.9,"maker":null},"model_count":3,"drop_count":1,"first_date":"2026-04-07","last_date":"2026-07-27","frontier_gain":7.8,"frontier":[{"date":"2026-04-07","model":"Claude Mythos Preview","score":83.1,"maker":"anthropic"},{"date":"2026-07-27","model":"Wiz Atlas","score":90.9,"maker":null}],"models":[{"model":"Wiz Atlas","score":90.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"MAI-Cyber-1-Flash","slug":"mai-cyber-1-flash","url":"https://handyai.news/modeldrop/mai-cyber-1-flash","published":"2026-07-27"},"launch_score":null,"reports":1,"maker":null,"open_weights":false,"released":"2026-07-27","dated_by":"first_cited","drop_url":null,"rank":1},{"model":"MAI-Cyber-1-Flash","score":86.26,"own_drop":true,"independent":true,"setting":"","source":{"drop":"MAI-Cyber-1-Flash","slug":"mai-cyber-1-flash","url":"https://handyai.news/modeldrop/mai-cyber-1-flash","published":"2026-07-27"},"launch_score":86.26,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-07-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-cyber-1-flash","rank":2},{"model":"Claude Mythos Preview","score":83.1,"own_drop":false,"independent":true,"setting":"","source":{"drop":"MAI-Cyber-1-Flash","slug":"mai-cyber-1-flash","url":"https://handyai.news/modeldrop/mai-cyber-1-flash","published":"2026-07-27"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":3}]},{"key":"cybersecurity-capture-the-flag-challenges","name":"Cybersecurity Capture The Flag Challenges","about":null,"type":{"key":"cyber","label":"Cyber"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#cybersecurity-capture-the-flag-challenges","api_url":"https://handyai.news/api/benchmarks/cybersecurity-capture-the-flag-challenges","image_url":"https://handyai.news/api/benchmarks/cybersecurity-capture-the-flag-challenges.png","leader":{"model":"GPT-5.3-Codex","score":77.6,"maker":"OpenAI"},"model_count":3,"drop_count":1,"first_date":"2025-12-11","last_date":"2026-02-05","frontier_gain":9.9,"frontier":[{"date":"2025-12-11","model":"GPT-5.2","score":67.7,"maker":"openai"},{"date":"2026-02-05","model":"GPT-5.3-Codex","score":77.6,"maker":"openai"}],"models":[{"model":"GPT-5.3-Codex","score":77.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.3-Codex","slug":"gpt-53-codex","url":"https://handyai.news/modeldrop/gpt-53-codex","published":"2026-02-05"},"launch_score":77.6,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-53-codex","rank":1},{"model":"GPT-5.2","score":67.7,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5.3-Codex","slug":"gpt-53-codex","url":"https://handyai.news/modeldrop/gpt-53-codex","published":"2026-02-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":2},{"model":"GPT-5.2-Codex","score":67.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.3-Codex","slug":"gpt-53-codex","url":"https://handyai.news/modeldrop/gpt-53-codex","published":"2026-02-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52-codex","rank":3}]},{"key":"dense200","name":"Dense200","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#dense200","api_url":"https://handyai.news/api/benchmarks/dense200","image_url":"https://handyai.news/api/benchmarks/dense200.png","leader":{"model":"Mistral Large 4","score":42,"maker":"Mistral AI"},"model_count":3,"drop_count":1,"first_date":"2026-07-16","last_date":"2026-10-06","frontier_gain":13.1,"frontier":[{"date":"2026-07-16","model":"Kimi K3","score":28.9,"maker":"moonshot"},{"date":"2026-09-03","model":"GPT-6 Astra","score":41.5,"maker":"openai"},{"date":"2026-10-06","model":"Mistral Large 4","score":42,"maker":"mistral"}],"models":[{"model":"Mistral Large 4","score":42,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":42,"reports":1,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":1},{"model":"GPT-6 Astra","score":41.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":2},{"model":"Kimi K3","score":28.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mistral Large 4","slug":"mistral-large-4","url":"https://handyai.news/modeldrop/mistral-large-4","published":"2026-10-06"},"launch_score":null,"reports":1,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":3}]},{"key":"frontiermath-tier-4","name":"FrontierMath Tier 4","about":null,"type":{"key":"math","label":"Math"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#frontiermath-tier-4","api_url":"https://handyai.news/api/benchmarks/frontiermath-tier-4","image_url":"https://handyai.news/api/benchmarks/frontiermath-tier-4.png","leader":{"model":"GPT-6 Astra","score":97.6,"maker":"OpenAI"},"model_count":3,"drop_count":1,"first_date":"2026-06-26","last_date":"2026-09-03","frontier_gain":14.6,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":83,"maker":"openai"},{"date":"2026-09-01","model":"Fable 5.1","score":87.8,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":97.6,"maker":"openai"}],"models":[{"model":"GPT-6 Astra","score":97.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-6 Astra","slug":"gpt-6-astra","url":"https://handyai.news/modeldrop/gpt-6-astra","published":"2026-09-03"},"launch_score":97.6,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":1},{"model":"Fable 5.1","score":87.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-6 Astra","slug":"gpt-6-astra","url":"https://handyai.news/modeldrop/gpt-6-astra","published":"2026-09-03"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":2},{"model":"GPT-5.6 Sol","score":83,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-6 Astra","slug":"gpt-6-astra","url":"https://handyai.news/modeldrop/gpt-6-astra","published":"2026-09-03"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":3}]},{"key":"gpqa-diamond-4-4-reliability","name":"GPQA Diamond 4/4 reliability","about":null,"type":{"key":"reasoning","label":"Reasoning"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#gpqa-diamond-4-4-reliability","api_url":"https://handyai.news/api/benchmarks/gpqa-diamond-4-4-reliability","image_url":"https://handyai.news/api/benchmarks/gpqa-diamond-4-4-reliability.png","leader":{"model":"o3-pro","score":76,"maker":"OpenAI"},"model_count":3,"drop_count":1,"first_date":"2025-04-16","last_date":"2025-06-10","frontier_gain":2,"frontier":[{"date":"2025-04-16","model":"o1-pro","score":74,"maker":"openai"},{"date":"2025-06-10","model":"o3-pro","score":76,"maker":"openai"}],"models":[{"model":"o3-pro","score":76,"own_drop":true,"independent":false,"setting":"","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":76,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-06-10","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-pro","rank":1},{"model":"o1-pro","score":74,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"first_cited","drop_url":null,"rank":2},{"model":"o3","score":67,"own_drop":false,"independent":false,"setting":"medium","source":{"drop":"o3-pro","slug":"o3-pro","url":"https://handyai.news/modeldrop/o3-pro","published":"2025-06-10"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":3}]},{"key":"hmmt-feb25","name":"HMMT Feb25","about":null,"type":{"key":"math","label":"Math"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#hmmt-feb25","api_url":"https://handyai.news/api/benchmarks/hmmt-feb25","image_url":"https://handyai.news/api/benchmarks/hmmt-feb25.png","leader":{"model":"Nemotron 3 Super","score":93.67,"maker":"NVIDIA"},"model_count":3,"drop_count":1,"first_date":"2025-08-05","last_date":"2026-03-11","frontier_gain":3.7,"frontier":[{"date":"2025-08-05","model":"gpt-oss-120b","score":90,"maker":"openai"},{"date":"2026-03-11","model":"Nemotron 3 Super","score":93.67,"maker":"nvidia"}],"models":[{"model":"Nemotron 3 Super","score":93.67,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":93.67,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-super","rank":1},{"model":"Qwen3.5-122B-A10B","score":91.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-03-11","dated_by":"first_cited","drop_url":null,"rank":2},{"model":"gpt-oss-120b","score":90,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Super","slug":"nemotron-3-super","url":"https://handyai.news/modeldrop/nemotron-3-super","published":"2026-03-11"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":3}]},{"key":"math","name":"MATH","about":null,"type":{"key":"math","label":"Math"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#math","api_url":"https://handyai.news/api/benchmarks/math","image_url":"https://handyai.news/api/benchmarks/math.png","leader":{"model":"Gemini 2.0 Flash","score":89.7,"maker":"Google DeepMind"},"model_count":3,"drop_count":1,"first_date":"2024-12-05","last_date":"2024-12-11","frontier_gain":3.2,"frontier":[{"date":"2024-12-05","model":"Gemini 1.5 Pro","score":86.5,"maker":"google"},{"date":"2024-12-11","model":"Gemini 2.0 Flash","score":89.7,"maker":"google"}],"models":[{"model":"Gemini 2.0 Flash","score":89.7,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":89.7,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":1},{"model":"Gemini 1.5 Pro","score":86.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":2},{"model":"Gemini 1.5 Flash","score":77.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"first_cited","drop_url":null,"rank":3}]},{"key":"mle-bench-30","name":"MLE-Bench-30","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#mle-bench-30","api_url":"https://handyai.news/api/benchmarks/mle-bench-30","image_url":"https://handyai.news/api/benchmarks/mle-bench-30.png","leader":{"model":"GPT-5.1-Codex-Max","score":17,"maker":"OpenAI"},"model_count":3,"drop_count":1,"first_date":"2025-11-19","last_date":"2025-12-18","frontier_gain":0,"frontier":[{"date":"2025-11-19","model":"GPT-5.1-Codex-Max","score":17,"maker":"openai"}],"models":[{"model":"GPT-5.1-Codex-Max","score":17,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51-codex-max","rank":1},{"model":"GPT-5.2","score":16,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52","rank":2},{"model":"GPT-5.2-Codex","score":10,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5.2-Codex","slug":"gpt-52-codex","url":"https://handyai.news/modeldrop/gpt-52-codex","published":"2025-12-18"},"launch_score":10,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-12-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-52-codex","rank":3}]},{"key":"mrcr-512k-1m","name":"MRCR 512K-1M","about":null,"type":{"key":"long-context","label":"Long context"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#mrcr-512k-1m","api_url":"https://handyai.news/api/benchmarks/mrcr-512k-1m","image_url":"https://handyai.news/api/benchmarks/mrcr-512k-1m.png","leader":{"model":"Muse Spark 1.3","score":98.1,"maker":"Meta"},"model_count":3,"drop_count":1,"first_date":"2026-06-26","last_date":"2026-09-02","frontier_gain":24.3,"frontier":[{"date":"2026-06-26","model":"GPT-5.6 Sol","score":73.8,"maker":"openai"},{"date":"2026-09-02","model":"Muse Spark 1.3","score":98.1,"maker":"meta"}],"models":[{"model":"Muse Spark 1.3","score":98.1,"own_drop":true,"independent":false,"setting":"max","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":98.1,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-13","rank":1},{"model":"GPT-5.6 Sol","score":73.8,"own_drop":false,"independent":false,"setting":"max","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-06-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-56-sol-terra-and-luna","rank":2},{"model":"Muse Spark 1.2","score":55.5,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"Muse Spark 1.3","slug":"muse-spark-13","url":"https://handyai.news/modeldrop/muse-spark-13","published":"2026-09-02"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-12","rank":3}]},{"key":"multichallenge","name":"MultiChallenge","about":null,"type":{"key":"instructions","label":"Instructions"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#multichallenge","api_url":"https://handyai.news/api/benchmarks/multichallenge","image_url":"https://handyai.news/api/benchmarks/multichallenge.png","leader":{"model":"GPT-4.5","score":43.8,"maker":"OpenAI"},"model_count":3,"drop_count":1,"first_date":"2024-12-05","last_date":"2025-04-14","frontier_gain":16,"frontier":[{"date":"2024-12-05","model":"GPT-4o","score":27.8,"maker":"openai"},{"date":"2025-02-27","model":"GPT-4.5","score":43.8,"maker":"openai"}],"models":[{"model":"GPT-4.5","score":43.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":1},{"model":"GPT-4.1","score":38.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":38.3,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":2},{"model":"GPT-4o","score":27.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":3}]},{"key":"natural2code","name":"Natural2Code","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#natural2code","api_url":"https://handyai.news/api/benchmarks/natural2code","image_url":"https://handyai.news/api/benchmarks/natural2code.png","leader":{"model":"Gemini 2.0 Flash","score":92.9,"maker":"Google DeepMind"},"model_count":3,"drop_count":1,"first_date":"2024-12-05","last_date":"2024-12-11","frontier_gain":7.5,"frontier":[{"date":"2024-12-05","model":"Gemini 1.5 Pro","score":85.4,"maker":"google"},{"date":"2024-12-11","model":"Gemini 2.0 Flash","score":92.9,"maker":"google"}],"models":[{"model":"Gemini 2.0 Flash","score":92.9,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":92.9,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-20-flash","rank":1},{"model":"Gemini 1.5 Pro","score":85.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":2},{"model":"Gemini 1.5 Flash","score":79.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Gemini 2.0 Flash","slug":"gemini-20-flash","url":"https://handyai.news/modeldrop/gemini-20-flash","published":"2024-12-11"},"launch_score":null,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2024-12-11","dated_by":"first_cited","drop_url":null,"rank":3}]},{"key":"openai-mrcr","name":"OpenAI-MRCR","about":null,"type":{"key":"long-context","label":"Long context"},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#openai-mrcr","api_url":"https://handyai.news/api/benchmarks/openai-mrcr","image_url":"https://handyai.news/api/benchmarks/openai-mrcr.png","leader":{"model":"GPT-4.5","score":38.5,"maker":"OpenAI"},"model_count":3,"drop_count":1,"first_date":"2024-12-05","last_date":"2025-04-14","frontier_gain":6.6,"frontier":[{"date":"2024-12-05","model":"GPT-4o","score":31.9,"maker":"openai"},{"date":"2025-02-27","model":"GPT-4.5","score":38.5,"maker":"openai"}],"models":[{"model":"GPT-4.5","score":38.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":1},{"model":"GPT-4o","score":31.9,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":2},{"model":"GPT-4.1","score":2,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":2,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":3}]},{"key":"swe-bench-openhands","name":"SWE-Bench (OpenHands)","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#swe-bench-openhands","api_url":"https://handyai.news/api/benchmarks/swe-bench-openhands","image_url":"https://handyai.news/api/benchmarks/swe-bench-openhands.png","leader":{"model":"Nemotron 3 Nano","score":38.8,"maker":"NVIDIA"},"model_count":3,"drop_count":1,"first_date":"2025-08-05","last_date":"2025-12-15","frontier_gain":4.8,"frontier":[{"date":"2025-08-05","model":"gpt-oss-20b","score":34,"maker":"openai"},{"date":"2025-12-15","model":"Nemotron 3 Nano","score":38.8,"maker":"nvidia"}],"models":[{"model":"Nemotron 3 Nano","score":38.8,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":38.8,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":1},{"model":"gpt-oss-20b","score":34,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":true,"released":"2025-08-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-oss","rank":2},{"model":"Qwen3-30B-A3B-Thinking-2507","score":22,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"first_cited","drop_url":null,"rank":3}]},{"key":"swe-bench-pro-v2-hard","name":"SWE-bench Pro v2 Hard","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#swe-bench-pro-v2-hard","api_url":"https://handyai.news/api/benchmarks/swe-bench-pro-v2-hard","image_url":"https://handyai.news/api/benchmarks/swe-bench-pro-v2-hard.png","leader":{"model":"Claude Opus 5.5","score":100,"maker":"Anthropic"},"model_count":3,"drop_count":1,"first_date":"2026-06-04","last_date":"2026-10-26","frontier_gain":68.6,"frontier":[{"date":"2026-06-04","model":"Nemotron 3 Ultra","score":31.4,"maker":"nvidia"},{"date":"2026-09-22","model":"Claude Opus 5.5","score":100,"maker":"anthropic"}],"models":[{"model":"Claude Opus 5.5","score":100,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mirendil, \"Introducing Fern\"","slug":"fern-01","url":"https://mirendil.com/news/introducing-fern/","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":1},{"model":"Fern 0.1","score":76.5,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Mirendil, \"Introducing Fern\"","slug":"fern-01","url":"https://mirendil.com/news/introducing-fern/","published":"2026-10-26"},"launch_score":76.5,"reports":1,"maker":{"key":"mirendil","name":"Mirendil","logo_url":"https://handyai.news/ui/makers/mirendil.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":null,"rank":2},{"model":"Nemotron 3 Ultra","score":31.4,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Mirendil, \"Introducing Fern\"","slug":"fern-01","url":"https://mirendil.com/news/introducing-fern/","published":"2026-10-26"},"launch_score":null,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":3}]},{"key":"swe-lancer-diamond","name":"SWE-Lancer Diamond","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#swe-lancer-diamond","api_url":"https://handyai.news/api/benchmarks/swe-lancer-diamond","image_url":"https://handyai.news/api/benchmarks/swe-lancer-diamond.png","leader":{"model":"GPT-4.5","score":32.6,"maker":"OpenAI"},"model_count":3,"drop_count":1,"first_date":"2024-12-05","last_date":"2025-02-27","frontier_gain":9.3,"frontier":[{"date":"2024-12-05","model":"GPT-4o","score":23.3,"maker":"openai"},{"date":"2025-02-27","model":"GPT-4.5","score":32.6,"maker":"openai"}],"models":[{"model":"GPT-4.5","score":32.6,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":32.6,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-02-27","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-45","rank":1},{"model":"GPT-4o","score":23.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":2},{"model":"o3-mini","score":10.8,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-4.5","slug":"gpt-45","url":"https://handyai.news/modeldrop/gpt-45","published":"2025-02-27"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-01-31","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-mini","rank":3}]},{"key":"bench-airline","name":"τ²-bench Airline","about":null,"type":{"key":"agents","label":"Agents"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#bench-airline","api_url":"https://handyai.news/api/benchmarks/bench-airline","image_url":"https://handyai.news/api/benchmarks/bench-airline.png","leader":{"model":"o3","score":64.8,"maker":"OpenAI"},"model_count":3,"drop_count":1,"first_date":"2025-04-14","last_date":"2025-08-07","frontier_gain":8.8,"frontier":[{"date":"2025-04-14","model":"GPT-4.1","score":56,"maker":"openai"},{"date":"2025-04-16","model":"o3","score":64.8,"maker":"openai"}],"models":[{"model":"o3","score":64.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o3-o4-mini","rank":1},{"model":"GPT-5","score":62.6,"own_drop":true,"independent":false,"setting":"high","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":62.6,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":2},{"model":"GPT-4.1","score":56,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-5","slug":"gpt-5","url":"https://handyai.news/modeldrop/gpt-5","published":"2025-08-07"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":3}]},{"key":"aa-omniscience","name":"AA-Omniscience","about":null,"type":{"key":"knowledge","label":"Knowledge"},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#aa-omniscience","api_url":"https://handyai.news/api/benchmarks/aa-omniscience","image_url":"https://handyai.news/api/benchmarks/aa-omniscience.png","leader":{"model":"Gemini 4 Carbon","score":54,"maker":"Google DeepMind"},"model_count":2,"drop_count":1,"first_date":"2026-09-22","last_date":"2026-10-28","frontier_gain":6,"frontier":[{"date":"2026-09-22","model":"Claude Opus 5.5","score":48,"maker":"anthropic"},{"date":"2026-10-28","model":"Gemini 4 Carbon","score":54,"maker":"google"}],"models":[{"model":"Gemini 4 Carbon","score":54,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":54,"reports":1,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":1},{"model":"Claude Opus 5.5","score":48,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Gemini 4 Carbon","slug":"gemini-4-carbon","url":"https://handyai.news/modeldrop/gemini-4-carbon","published":"2026-10-28"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":2}]},{"key":"artifacts-bench","name":"Artifacts Bench","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#artifacts-bench","api_url":"https://handyai.news/api/benchmarks/artifacts-bench","image_url":"https://handyai.news/api/benchmarks/artifacts-bench.png","leader":{"model":"Claude Haiku 4.5","score":36.6,"maker":"Anthropic"},"model_count":2,"drop_count":1,"first_date":"2025-10-15","last_date":"2026-06-02","frontier_gain":0,"frontier":[{"date":"2025-10-15","model":"Claude Haiku 4.5","score":36.6,"maker":"anthropic"}],"models":[{"model":"Claude Haiku 4.5","score":36.6,"own_drop":false,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2025-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-45","rank":1},{"model":"MAI-Code-1-Flash","score":36.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"MAI-Code-1-Flash","slug":"mai-code-1-flash","url":"https://handyai.news/modeldrop/mai-code-1-flash","published":"2026-06-02"},"launch_score":36.4,"reports":1,"maker":{"key":"microsoft","name":"Microsoft","logo_url":"https://handyai.news/ui/makers/microsoft.svg"},"open_weights":false,"released":"2026-06-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mai-code-1-flash","rank":2}]},{"key":"bfcl-v2-live","name":"BFCL V2 Live","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#bfcl-v2-live","api_url":"https://handyai.news/api/benchmarks/bfcl-v2-live","image_url":"https://handyai.news/api/benchmarks/bfcl-v2-live.png","leader":{"model":"Llama 3.1 Nemotron Ultra","score":74.1,"maker":"NVIDIA"},"model_count":2,"drop_count":1,"first_date":"2024-12-26","last_date":"2025-04-07","frontier_gain":15.4,"frontier":[{"date":"2024-12-26","model":"Llama 3.1 405B","score":58.7,"maker":"meta"},{"date":"2025-04-07","model":"Llama 3.1 Nemotron Ultra","score":74.1,"maker":"nvidia"}],"models":[{"model":"Llama 3.1 Nemotron Ultra","score":74.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Llama 3.1 Nemotron Ultra","slug":"llama-31-nemotron-ultra","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","published":"2025-04-07"},"launch_score":74.1,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","rank":1},{"model":"Llama 3.1 405B","score":58.7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Llama 3.1 Nemotron Ultra","slug":"llama-31-nemotron-ultra","url":"https://handyai.news/modeldrop/llama-31-nemotron-ultra","published":"2025-04-07"},"launch_score":null,"reports":1,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2024-12-26","dated_by":"first_cited","drop_url":null,"rank":2}]},{"key":"code-refactoring-tasks","name":"Code refactoring tasks","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#code-refactoring-tasks","api_url":"https://handyai.news/api/benchmarks/code-refactoring-tasks","image_url":"https://handyai.news/api/benchmarks/code-refactoring-tasks.png","leader":{"model":"GPT-5-Codex","score":51.3,"maker":"OpenAI"},"model_count":2,"drop_count":1,"first_date":"2025-08-07","last_date":"2025-09-15","frontier_gain":17.4,"frontier":[{"date":"2025-08-07","model":"GPT-5","score":33.9,"maker":"openai"},{"date":"2025-09-15","model":"GPT-5-Codex","score":51.3,"maker":"openai"}],"models":[{"model":"GPT-5-Codex","score":51.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5-Codex","slug":"gpt-5-codex","url":"https://handyai.news/modeldrop/gpt-5-codex","published":"2025-09-15"},"launch_score":51.3,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-09-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5-codex","rank":1},{"model":"GPT-5","score":33.9,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5-Codex","slug":"gpt-5-codex","url":"https://handyai.news/modeldrop/gpt-5-codex","published":"2025-09-15"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":2}]},{"key":"code-review-comments-per-pr","name":"Code review comments per PR","about":null,"type":{"key":"coding","label":"Coding"},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#code-review-comments-per-pr","api_url":"https://handyai.news/api/benchmarks/code-review-comments-per-pr","image_url":"https://handyai.news/api/benchmarks/code-review-comments-per-pr.png","leader":{"model":"GPT-5","score":1.32,"maker":"OpenAI"},"model_count":2,"drop_count":1,"first_date":"2025-08-07","last_date":"2025-09-15","frontier_gain":0,"frontier":[{"date":"2025-08-07","model":"GPT-5","score":1.32,"maker":"openai"}],"models":[{"model":"GPT-5","score":1.32,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5-Codex","slug":"gpt-5-codex","url":"https://handyai.news/modeldrop/gpt-5-codex","published":"2025-09-15"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":1},{"model":"GPT-5-Codex","score":0.93,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5-Codex","slug":"gpt-5-codex","url":"https://handyai.news/modeldrop/gpt-5-codex","published":"2025-09-15"},"launch_score":0.93,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-09-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5-codex","rank":2}]},{"key":"code-review-high-impact-comments","name":"Code review high-impact comments","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#code-review-high-impact-comments","api_url":"https://handyai.news/api/benchmarks/code-review-high-impact-comments","image_url":"https://handyai.news/api/benchmarks/code-review-high-impact-comments.png","leader":{"model":"GPT-5-Codex","score":52.4,"maker":"OpenAI"},"model_count":2,"drop_count":1,"first_date":"2025-08-07","last_date":"2025-09-15","frontier_gain":13,"frontier":[{"date":"2025-08-07","model":"GPT-5","score":39.4,"maker":"openai"},{"date":"2025-09-15","model":"GPT-5-Codex","score":52.4,"maker":"openai"}],"models":[{"model":"GPT-5-Codex","score":52.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5-Codex","slug":"gpt-5-codex","url":"https://handyai.news/modeldrop/gpt-5-codex","published":"2025-09-15"},"launch_score":52.4,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-09-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5-codex","rank":1},{"model":"GPT-5","score":39.4,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5-Codex","slug":"gpt-5-codex","url":"https://handyai.news/modeldrop/gpt-5-codex","published":"2025-09-15"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":2}]},{"key":"code-review-incorrect-comments","name":"Code review incorrect comments","about":null,"type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":true,"url":"https://handyai.news/benchmarks#code-review-incorrect-comments","api_url":"https://handyai.news/api/benchmarks/code-review-incorrect-comments","image_url":"https://handyai.news/api/benchmarks/code-review-incorrect-comments.png","leader":{"model":"GPT-5-Codex","score":4.4,"maker":"OpenAI"},"model_count":2,"drop_count":1,"first_date":"2025-08-07","last_date":"2025-09-15","frontier_gain":9.3,"frontier":[{"date":"2025-08-07","model":"GPT-5","score":13.7,"maker":"openai"},{"date":"2025-09-15","model":"GPT-5-Codex","score":4.4,"maker":"openai"}],"models":[{"model":"GPT-5-Codex","score":4.4,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-5-Codex","slug":"gpt-5-codex","url":"https://handyai.news/modeldrop/gpt-5-codex","published":"2025-09-15"},"launch_score":4.4,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-09-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5-codex","rank":1},{"model":"GPT-5","score":13.7,"own_drop":false,"independent":false,"setting":"high","source":{"drop":"GPT-5-Codex","slug":"gpt-5-codex","url":"https://handyai.news/modeldrop/gpt-5-codex","published":"2025-09-15"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":2}]},{"key":"factscore","name":"FActScore","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":true,"url":"https://handyai.news/benchmarks#factscore","api_url":"https://handyai.news/api/benchmarks/factscore","image_url":"https://handyai.news/api/benchmarks/factscore.png","leader":{"model":"Grok 4.1","score":2.97,"maker":"SpaceXAI"},"model_count":2,"drop_count":1,"first_date":"2025-09-19","last_date":"2025-11-17","frontier_gain":6.9,"frontier":[{"date":"2025-09-19","model":"Grok 4 Fast","score":9.89,"maker":"xai"},{"date":"2025-11-17","model":"Grok 4.1","score":2.97,"maker":"xai"}],"models":[{"model":"Grok 4.1","score":2.97,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":2.97,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-11-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-41","rank":1},{"model":"Grok 4 Fast","score":9.89,"own_drop":false,"independent":false,"setting":"non reasoning","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-09-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4-fast","rank":2}]},{"key":"hallucination-rate","name":"Hallucination Rate","about":null,"type":{"key":"knowledge","label":"Knowledge"},"unit":"%","lower_is_better":true,"url":"https://handyai.news/benchmarks#hallucination-rate","api_url":"https://handyai.news/api/benchmarks/hallucination-rate","image_url":"https://handyai.news/api/benchmarks/hallucination-rate.png","leader":{"model":"Grok 4.1","score":4.22,"maker":"SpaceXAI"},"model_count":2,"drop_count":1,"first_date":"2025-09-19","last_date":"2025-11-17","frontier_gain":7.9,"frontier":[{"date":"2025-09-19","model":"Grok 4 Fast","score":12.09,"maker":"xai"},{"date":"2025-11-17","model":"Grok 4.1","score":4.22,"maker":"xai"}],"models":[{"model":"Grok 4.1","score":4.22,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":4.22,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-11-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-41","rank":1},{"model":"Grok 4 Fast","score":12.09,"own_drop":false,"independent":false,"setting":"non reasoning","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-09-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4-fast","rank":2}]},{"key":"hermes-bench","name":"Hermes Bench","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#hermes-bench","api_url":"https://handyai.news/api/benchmarks/hermes-bench","image_url":"https://handyai.news/api/benchmarks/hermes-bench.png","leader":{"model":"Nemotron 3.5 Super VL","score":63.44,"maker":"NVIDIA"},"model_count":2,"drop_count":1,"first_date":"2026-06-04","last_date":"2026-10-15","frontier_gain":7.1,"frontier":[{"date":"2026-06-04","model":"Nemotron 3 Ultra","score":56.33,"maker":"nvidia"},{"date":"2026-10-15","model":"Nemotron 3.5 Super VL","score":63.44,"maker":"nvidia"}],"models":[{"model":"Nemotron 3.5 Super VL","score":63.44,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":63.44,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":1},{"model":"Nemotron 3 Ultra","score":56.33,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3.5 Super VL","slug":"nemotron-35-super-vl","url":"https://handyai.news/modeldrop/nemotron-35-super-vl","published":"2026-10-15"},"launch_score":null,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-06-04","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-ultra","rank":2}]},{"key":"irregular-medium-cyber-challenges","name":"Irregular medium cyber challenges","about":null,"type":{"key":"cyber","label":"Cyber"},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#irregular-medium-cyber-challenges","api_url":"https://handyai.news/api/benchmarks/irregular-medium-cyber-challenges","image_url":"https://handyai.news/api/benchmarks/irregular-medium-cyber-challenges.png","leader":{"model":"GPT-5","score":11,"maker":"OpenAI"},"model_count":2,"drop_count":1,"first_date":"2025-08-07","last_date":"2025-11-19","frontier_gain":0,"frontier":[{"date":"2025-08-07","model":"GPT-5","score":11,"maker":"openai"}],"models":[{"model":"GPT-5","score":11,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.1-Codex-Max","slug":"gpt-51-codex-max","url":"https://handyai.news/modeldrop/gpt-51-codex-max","published":"2025-11-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-08-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-5","rank":1},{"model":"GPT-5.1-Codex-Max","score":9,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.1-Codex-Max","slug":"gpt-51-codex-max","url":"https://handyai.news/modeldrop/gpt-51-codex-max","published":"2025-11-19"},"launch_score":9,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51-codex-max","rank":2}]},{"key":"mgsm","name":"MGSM","about":null,"type":{"key":"math","label":"Math"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#mgsm","api_url":"https://handyai.news/api/benchmarks/mgsm","image_url":"https://handyai.news/api/benchmarks/mgsm.png","leader":{"model":"o1-preview","score":90.8,"maker":"OpenAI"},"model_count":2,"drop_count":1,"first_date":"2024-12-05","last_date":"2024-12-05","frontier_gain":0,"frontier":[{"date":"2024-12-05","model":"o1-preview","score":90.8,"maker":"openai"}],"models":[{"model":"o1-preview","score":90.8,"own_drop":false,"independent":false,"setting":"","source":{"drop":"o1","slug":"o1","url":"https://handyai.news/modeldrop/o1","published":"2024-12-05"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":1},{"model":"o1","score":89.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"o1","slug":"o1","url":"https://handyai.news/modeldrop/o1","published":"2024-12-05"},"launch_score":89.3,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/o1","rank":2}]},{"key":"openai-prs","name":"OpenAI PRs","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#openai-prs","api_url":"https://handyai.news/api/benchmarks/openai-prs","image_url":"https://handyai.news/api/benchmarks/openai-prs.png","leader":{"model":"GPT-5.1-Codex-Max","score":53,"maker":"OpenAI"},"model_count":2,"drop_count":1,"first_date":"2025-11-12","last_date":"2025-11-19","frontier_gain":8,"frontier":[{"date":"2025-11-12","model":"GPT-5.1","score":45,"maker":"openai"},{"date":"2025-11-19","model":"GPT-5.1-Codex-Max","score":53,"maker":"openai"}],"models":[{"model":"GPT-5.1-Codex-Max","score":53,"own_drop":true,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.1-Codex-Max","slug":"gpt-51-codex-max","url":"https://handyai.news/modeldrop/gpt-51-codex-max","published":"2025-11-19"},"launch_score":53,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51-codex-max","rank":1},{"model":"GPT-5.1","score":45,"own_drop":false,"independent":false,"setting":"xhigh","source":{"drop":"GPT-5.1-Codex-Max","slug":"gpt-51-codex-max","url":"https://handyai.news/modeldrop/gpt-51-codex-max","published":"2025-11-19"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-11-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-51","rank":2}]},{"key":"ruler-128k","name":"RULER (128K)","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#ruler-128k","api_url":"https://handyai.news/api/benchmarks/ruler-128k","image_url":"https://handyai.news/api/benchmarks/ruler-128k.png","leader":{"model":"Nemotron Nano 2","score":78.9,"maker":"NVIDIA"},"model_count":2,"drop_count":1,"first_date":"2025-08-18","last_date":"2025-08-18","frontier_gain":0,"frontier":[{"date":"2025-08-18","model":"Nemotron Nano 2","score":78.9,"maker":"nvidia"}],"models":[{"model":"Nemotron Nano 2","score":78.9,"own_drop":true,"independent":false,"setting":"reasoning off","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":78.9,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-nano-2","rank":1},{"model":"Qwen3-8B","score":74.1,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron Nano 2","slug":"nemotron-nano-2","url":"https://handyai.news/modeldrop/nemotron-nano-2","published":"2025-08-18"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-08-18","dated_by":"first_cited","drop_url":null,"rank":2}]},{"key":"ruler-100-1m","name":"RULER-100@1M","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#ruler-100-1m","api_url":"https://handyai.news/api/benchmarks/ruler-100-1m","image_url":"https://handyai.news/api/benchmarks/ruler-100-1m.png","leader":{"model":"Nemotron 3 Nano","score":86.3,"maker":"NVIDIA"},"model_count":2,"drop_count":1,"first_date":"2025-12-15","last_date":"2025-12-15","frontier_gain":0,"frontier":[{"date":"2025-12-15","model":"Nemotron 3 Nano","score":86.3,"maker":"nvidia"}],"models":[{"model":"Nemotron 3 Nano","score":86.3,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":86.3,"reports":1,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-3-nano","rank":1},{"model":"Qwen3-30B-A3B-Thinking-2507","score":77.5,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Nemotron 3 Nano","slug":"nemotron-3-nano","url":"https://handyai.news/modeldrop/nemotron-3-nano","published":"2025-12-15"},"launch_score":null,"reports":1,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2025-12-15","dated_by":"first_cited","drop_url":null,"rank":2}]},{"key":"sycophancy-rate","name":"Sycophancy rate","about":null,"type":{"key":"other","label":"Other"},"unit":"%","lower_is_better":true,"url":"https://handyai.news/benchmarks#sycophancy-rate","api_url":"https://handyai.news/api/benchmarks/sycophancy-rate","image_url":"https://handyai.news/api/benchmarks/sycophancy-rate.png","leader":{"model":"Grok 4","score":7,"maker":"SpaceXAI"},"model_count":2,"drop_count":1,"first_date":"2025-07-09","last_date":"2025-11-17","frontier_gain":0,"frontier":[{"date":"2025-07-09","model":"Grok 4","score":7,"maker":"xai"}],"models":[{"model":"Grok 4","score":7,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":1},{"model":"Grok 4.1","score":19,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok 4.1","slug":"grok-41","url":"https://handyai.news/modeldrop/grok-41","published":"2025-11-17"},"launch_score":19,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-11-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-41","rank":2}]},{"key":"uk-aisi-the-last-ones-cyber-range","name":"UK AISI \"The Last Ones\" cyber range","about":null,"type":{"key":"cyber","label":"Cyber"},"unit":null,"lower_is_better":false,"url":"https://handyai.news/benchmarks#uk-aisi-the-last-ones-cyber-range","api_url":"https://handyai.news/api/benchmarks/uk-aisi-the-last-ones-cyber-range","image_url":"https://handyai.news/api/benchmarks/uk-aisi-the-last-ones-cyber-range.png","leader":{"model":"Claude Mythos Preview","score":22,"maker":"Anthropic"},"model_count":2,"drop_count":1,"first_date":"2026-02-05","last_date":"2026-04-07","frontier_gain":6,"frontier":[{"date":"2026-02-05","model":"Claude Opus 4.6","score":16,"maker":"anthropic"},{"date":"2026-04-07","model":"Claude Mythos Preview","score":22,"maker":"anthropic"}],"models":[{"model":"Claude Mythos Preview","score":22,"own_drop":true,"independent":true,"setting":"","source":{"drop":"Claude Mythos Preview","slug":"claude-mythos-preview","url":"https://handyai.news/modeldrop/claude-mythos-preview","published":"2026-04-07"},"launch_score":22,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-04-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-mythos-preview","rank":1},{"model":"Claude Opus 4.6","score":16,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Claude Mythos Preview","slug":"claude-mythos-preview","url":"https://handyai.news/modeldrop/claude-mythos-preview","published":"2026-04-07"},"launch_score":null,"reports":1,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-02-05","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-46","rank":2}]},{"key":"video-mme","name":"Video-MME","about":null,"type":{"key":"multimodal","label":"Multimodal"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#video-mme","api_url":"https://handyai.news/api/benchmarks/video-mme","image_url":"https://handyai.news/api/benchmarks/video-mme.png","leader":{"model":"GPT-4.1","score":72,"maker":"OpenAI"},"model_count":2,"drop_count":1,"first_date":"2024-12-05","last_date":"2025-04-14","frontier_gain":6.7,"frontier":[{"date":"2024-12-05","model":"GPT-4o","score":65.3,"maker":"openai"},{"date":"2025-04-14","model":"GPT-4.1","score":72,"maker":"openai"}],"models":[{"model":"GPT-4.1","score":72,"own_drop":true,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":72,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2025-04-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-41","rank":1},{"model":"GPT-4o","score":65.3,"own_drop":false,"independent":false,"setting":"","source":{"drop":"GPT-4.1","slug":"gpt-41","url":"https://handyai.news/modeldrop/gpt-41","published":"2025-04-14"},"launch_score":null,"reports":1,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2024-12-05","dated_by":"first_cited","drop_url":null,"rank":2}]},{"key":"wmdp-cyber","name":"WMDP Cyber","about":null,"type":{"key":"cyber","label":"Cyber"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#wmdp-cyber","api_url":"https://handyai.news/api/benchmarks/wmdp-cyber","image_url":"https://handyai.news/api/benchmarks/wmdp-cyber.png","leader":{"model":"Grok 4","score":79,"maker":"SpaceXAI"},"model_count":2,"drop_count":1,"first_date":"2025-07-09","last_date":"2025-08-28","frontier_gain":0,"frontier":[{"date":"2025-07-09","model":"Grok 4","score":79,"maker":"xai"}],"models":[{"model":"Grok 4","score":79,"own_drop":false,"independent":false,"setting":"","source":{"drop":"Grok Code Fast 1","slug":"grok-code-fast-1","url":"https://handyai.news/modeldrop/grok-code-fast-1","published":"2025-08-28"},"launch_score":null,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-07-09","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-4","rank":1},{"model":"Grok Code Fast 1","score":62.1,"own_drop":true,"independent":false,"setting":"","source":{"drop":"Grok Code Fast 1","slug":"grok-code-fast-1","url":"https://handyai.news/modeldrop/grok-code-fast-1","published":"2025-08-28"},"launch_score":62.1,"reports":1,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2025-08-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-code-fast-1","rank":2}]},{"key":"critpt-v2-hard","name":"CritPt v2 Hard","about":"Unpublished research-level physics problems, hard subset, run by Artificial Analysis.","type":{"key":"reasoning","label":"Reasoning"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#critpt-v2-hard","api_url":"https://handyai.news/api/benchmarks/critpt-v2-hard","image_url":"https://handyai.news/api/benchmarks/critpt-v2-hard.png","leader":{"model":"GPT-6 Astra","score":70.5,"maker":"OpenAI"},"model_count":27,"drop_count":0,"first_date":"2026-07-15","last_date":"2026-10-28","frontier_gain":52.5,"frontier":[{"date":"2026-07-15","model":"Inkling","score":18,"maker":"thinkingmachines"},{"date":"2026-07-16","model":"Kimi K3","score":40.5,"maker":"moonshot"},{"date":"2026-09-01","model":"Fable 5.1","score":52.5,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":70.5,"maker":"openai"}],"models":[{"model":"GPT-6 Astra","score":70.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":1},{"model":"GPT-6.1 Sol","score":69.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":2},{"model":"Gemini 4 Carbon","score":69,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":3},{"model":"GPT-6.1 Terra","score":68,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":4},{"model":"Claude Opus 5.5","score":68,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":5},{"model":"Claude Sonnet 5.5","score":59,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":6},{"model":"Muse Wonder Full","score":56.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":7},{"model":"Gemini 4 Argon","score":52.5,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":8},{"model":"Fable 5.1","score":52.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":9},{"model":"Muse Wonder","score":50.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":10},{"model":"Kimi K3.1 Pro","score":49.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":11},{"model":"Gemini 3.8 Flash","score":46.5,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":12},{"model":"Claude Haiku 5.5","score":43.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":13},{"model":"GPT-6 Luna","score":42.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":14},{"model":"Kimi K3","score":40.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":15},{"model":"Ling-3.1-flash-1019","score":38.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"antgroup","name":"Ant Group","logo_url":"https://handyai.news/ui/makers/antgroup.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":16},{"model":"MiMo-V2.6-Pro","score":34,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":17},{"model":"GLM-5.3","score":33.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":18},{"model":"Grok 4.7","score":33.5,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":19},{"model":"MiniMax M3.1 Flash","score":32.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":20},{"model":"Qwen3.8-Max","score":29.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":21},{"model":"Muse Spark Open","score":29.5,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":22},{"model":"Mistral Large 4","score":27,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":23},{"model":"DeepSeek V4.1 Flash 1017","score":25,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":24},{"model":"Inkling","score":18,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":25},{"model":"Nemotron 3.5 Super VL","score":13.5,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":26},{"model":"Qwen3.8","score":4,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":27}]},{"key":"terminal-bench-science-standard-harness","name":"Terminal-Bench Science (standard harness)","about":"Terminal-Bench Science run by a third party on one shared harness.","type":{"key":"coding","label":"Coding"},"unit":"%","lower_is_better":false,"url":"https://handyai.news/benchmarks#terminal-bench-science-standard-harness","api_url":"https://handyai.news/api/benchmarks/terminal-bench-science-standard-harness","image_url":"https://handyai.news/api/benchmarks/terminal-bench-science-standard-harness.png","leader":{"model":"Gemini 4 Carbon","score":64.8,"maker":"Google DeepMind"},"model_count":27,"drop_count":0,"first_date":"2026-07-15","last_date":"2026-10-28","frontier_gain":64.8,"frontier":[{"date":"2026-07-15","model":"Inkling","score":0,"maker":"thinkingmachines"},{"date":"2026-07-16","model":"Kimi K3","score":8.1,"maker":"moonshot"},{"date":"2026-07-19","model":"Qwen3.8-Max","score":11.9,"maker":"qwen"},{"date":"2026-09-01","model":"Fable 5.1","score":43.3,"maker":"anthropic"},{"date":"2026-09-03","model":"GPT-6 Astra","score":63.3,"maker":"openai"},{"date":"2026-10-28","model":"Gemini 4 Carbon","score":64.8,"maker":"google"}],"models":[{"model":"Gemini 4 Carbon","score":64.8,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-carbon","rank":1},{"model":"GPT-6 Astra","score":63.3,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-03","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-astra","rank":2},{"model":"Claude Opus 5.5","score":59,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-opus-55","rank":3},{"model":"GPT-6.1 Sol","score":58.1,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-29","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-sol","rank":4},{"model":"GPT-6.1 Terra","score":57.6,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-10-26","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-61-terra","rank":5},{"model":"Claude Sonnet 5.5","score":53.3,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-28","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-sonnet-55","rank":6},{"model":"Gemini 4 Argon","score":52.9,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-10-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-4-argon","rank":7},{"model":"Fable 5.1","score":43.3,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-09-01","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/fable-51","rank":8},{"model":"Muse Wonder Full","score":37.1,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":9},{"model":"Kimi K3.1 Pro","score":34.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-10-23","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k31-pro","rank":10},{"model":"Muse Wonder","score":25.7,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":false,"released":"2026-10-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-wonder","rank":11},{"model":"Claude Haiku 5.5","score":20,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"anthropic","name":"Anthropic","logo_url":"https://handyai.news/ui/makers/anthropic.svg"},"open_weights":false,"released":"2026-10-07","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/claude-haiku-55","rank":12},{"model":"DeepSeek V4.1 Flash 1017","score":12.9,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"deepseek","name":"DeepSeek","logo_url":"https://handyai.news/ui/makers/deepseek.svg"},"open_weights":true,"released":"2026-10-17","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/deepseek-v41-flash-1017","rank":13},{"model":"Qwen3.8-Max","score":11.9,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":false,"released":"2026-07-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38-max","rank":14},{"model":"Grok 4.7","score":11.4,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xai","name":"SpaceXAI","logo_url":"https://handyai.news/ui/makers/xai.svg"},"open_weights":false,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/grok-47","rank":15},{"model":"Gemini 3.8 Flash","score":10,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"google","name":"Google DeepMind","logo_url":"https://handyai.news/ui/makers/google.svg"},"open_weights":false,"released":"2026-09-02","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gemini-38-flash","rank":16},{"model":"GLM-5.3","score":9.5,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"zai","name":"Z.ai","logo_url":"https://handyai.news/ui/makers/zai.svg"},"open_weights":true,"released":"2026-08-14","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/glm-53","rank":17},{"model":"GPT-6 Luna","score":8.6,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"openai","name":"OpenAI","logo_url":"https://handyai.news/ui/makers/openai.svg"},"open_weights":false,"released":"2026-09-22","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/gpt-6-sol-and-gpt-6-luna","rank":18},{"model":"Kimi K3","score":8.1,"own_drop":false,"independent":true,"setting":"max","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"moonshot","name":"Moonshot AI","logo_url":"https://handyai.news/ui/makers/moonshot.svg"},"open_weights":true,"released":"2026-07-16","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/kimi-k3","rank":19},{"model":"MiMo-V2.6-Pro","score":5.7,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"xiaomi","name":"Xiaomi","logo_url":"https://handyai.news/ui/makers/xiaomi.svg"},"open_weights":true,"released":"2026-09-21","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mimo-v26-pro","rank":20},{"model":"Mistral Large 4","score":4.3,"own_drop":false,"independent":true,"setting":"high","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"mistral","name":"Mistral AI","logo_url":"https://handyai.news/ui/makers/mistral.svg"},"open_weights":true,"released":"2026-10-06","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/mistral-large-4","rank":21},{"model":"Ling-3.1-flash-1019","score":4.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"antgroup","name":"Ant Group","logo_url":"https://handyai.news/ui/makers/antgroup.svg"},"open_weights":true,"released":"2026-10-27","dated_by":"own_drop","drop_url":null,"rank":22},{"model":"MiniMax M3.1 Flash","score":3.3,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"minimax","name":"MiniMax","logo_url":"https://handyai.news/ui/makers/minimax.svg"},"open_weights":true,"released":"2026-10-19","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/minimax-m31-flash","rank":23},{"model":"Muse Spark Open","score":1.9,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"meta","name":"Meta","logo_url":"https://handyai.news/ui/makers/meta.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/muse-spark-open","rank":24},{"model":"Qwen3.8","score":0.5,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"qwen","name":"Alibaba","logo_url":"https://handyai.news/ui/makers/qwen.svg"},"open_weights":true,"released":"2026-08-12","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/qwen38","rank":25},{"model":"Nemotron 3.5 Super VL","score":0,"own_drop":false,"independent":true,"setting":"","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"nvidia","name":"NVIDIA","logo_url":"https://handyai.news/ui/makers/nvidia.svg"},"open_weights":true,"released":"2026-10-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/nemotron-35-super-vl","rank":26},{"model":"Inkling","score":0,"own_drop":false,"independent":true,"setting":"xhigh","source":{"drop":"Artificial Analysis (2026-11-04)","slug":"aa-auto-2026-11-04","url":"https://artificialanalysis.ai/evaluations","published":"2026-11-04"},"launch_score":null,"reports":2,"maker":{"key":"thinkingmachines","name":"Thinking Machines","logo_url":"https://handyai.news/ui/makers/thinkingmachines.svg"},"open_weights":true,"released":"2026-07-15","dated_by":"own_drop","drop_url":"https://handyai.news/modeldrop/inkling","rank":27}]}],"range":{"key":"all","label":"All time","start":null}}