---
title: Handy AI Sentiment Bench
url: https://handyai.news/benchmarks/sentiment-bench
generated_at: 2026-11-04T17:21:10.777Z
method: v2
---

# Handy AI Sentiment Bench

> Handy AI Sentiment Bench: how people received every major AI model in its first week, scored from X, Hacker News, the news and trusted reviewers, and whether they say it lives up to its benchmarks.

Points out of 100 for how a model was received in its first seven days (from its launch: the biggest Hacker News launch thread near the drop date, or the drop date). Claude Haiku 5.5 reads up to 60 of the most-engaged posts on X (each checked against X), 250 Hacker News comments in the order readers voted them, 60 news headlines and the week's write-ups from a short list of trusted independent reviewers, and judges each against the model: its stance, what it rests on (own use, a cited test, opinion, passing on the announcement, or talk about the company), and whether it beats, matches or falls short of its benchmarks and launch claims. Own use counts double, a cited test 1.5 times, passing on the announcement 0.3 times, and remarks about the company, its policies or the launch event not at all. The maker's accounts, staff, partners, resellers and press releases are left out. Each crowd source is measured against its own norm (Hacker News runs far colder than X), then X, Hacker News and the news blend 45/35/20, a source with too little to go on counting as typical. Trusted reviewers (Every, Simon Willison, Zvi Mowshowitz, Ethan Mollick, Nathan Lambert and others) get one vote each and take 30% of reception when any of them weighed in. A quarter of the score is whether people who used it, or cite a test, say it lives up to its benchmarks, with reviewers' verdicts counting triple. Full method: https://handyai.news/benchmarks/sentiment-bench#method

| Rank | Model | Maker | Score | Reception | Lives up to its numbers | X | Hacker News | News | Trusted reviewers | Opinions | Launched |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | Claude Opus 4.5 | Anthropic | 91.5 | 94.2 | 83.3 | 93.5 | 63.2 | 97.1 | 91.5 (Every 100, Ethan Mollick 100, Simon Willison 100, Zvi Mowshowitz 100) | 129 | 2025-11-24 |
| 2 | GLM 5.2 | Z.ai | 88 | 91.2 | 78.5 | 89.9 | 68.4 | 97.7 | 78.5 (Nathan Lambert 100, Zvi Mowshowitz 75, Sebastian Raschka 75) | 180 | 2026-06-20 |
| 3 | Grok 4.6 | SpaceXAI | 87 | 89.2 | 80.2 | 89.8 | 61.6 | 97.8 |  | 155 | 2026-08-12 |
| 4 | Gemini 4 Carbon | Google DeepMind | 85.8 | 85.2 | 87.8 | 89 | 47 | 95.5 | 81.3 (Peter Gostev 100) | 98 | 2026-10-28 |
| 5 | GPT-5.3-Codex | OpenAI | 85.6 | 86.7 | 82.1 | 88.4 | 58.3 | 94.3 | 78.5 (Max Woolf 100, Simon Willison 75, Every 75) | 107 | 2026-02-05 |
| 6 | Claude Opus 5.5 | Anthropic | 85.2 | 86.5 | 81.6 | 92.3 | 59.6 | 90.7 | 75.8 (Ethan Mollick 100, Every 100, Zvi Mowshowitz 90, Simon Willison 75, Max Woolf 75, xlr8harder 25) | 204 | 2026-09-22 |
| 7 | Gemini 3 Flash | Google DeepMind | 84.5 | 86.1 | 79.6 | 78.2 | 69.5 | 96.6 | 77.2 (Peter Gostev 92, Simon Willison 75) | 156 | 2025-12-17 |
| 8 | Composer | Cursor | 84.1 | 83.8 | 84.8 | 90.8 | 61.5 |  | 71.3 (Every 75) | 72 | 2025-10-29 |
| 9 | DeepSeek V4.1 Flash 1017 | DeepSeek | 83.9 | 84.2 | 83 | 86.4 | 82.6 | 94.9 | 67.4 (Nathan Lambert 100, Simon Willison 75, xlr8harder 25) | 143 | 2026-10-17 |
| 10 | Kimi K2.5 | Moonshot AI | 83.7 | 85.5 | 78.3 | 84 | 83.1 | 96.1 | 71.3 (Simon Willison 75) | 135 | 2026-01-27 |
| 11 | Qwen3.8 | Alibaba | 83.6 | 87.7 | 71.4 | 81.3 | 73.2 | 96.6 | 77.8 (Simon Willison 91) | 218 | 2026-08-12 |
| 12 | Gemini 3 Pro | Google DeepMind | 82.9 | 87.9 | 68 | 83.8 | 63.7 | 94.1 | 81.3 (Andrej Karpathy 100, Ethan Mollick 100, Simon Willison 75, Every 75, Zvi Mowshowitz 75) | 179 | 2025-11-18 |
| 13 | Grok 4.5 | SpaceXAI | 82.5 | 81.8 | 84.7 | 99.2 | 58.5 | 91 | 61.8 (Sebastian Raschka 75, Max Woolf 75, Ethan Mollick 25) | 145 | 2026-07-09 |
| 14 | Kimi K2.7 Code | Moonshot AI | 82.5 | 84.1 | 77.8 | 79.6 | 71.2 |  |  | 73 | 2026-06-12 |
| 15 | GPT-5.5 | OpenAI | 82.5 | 84 | 78.2 | 85.1 | 50.7 | 95.1 | 79.8 (Peter Gostev 100, Ethan Mollick 100, Every 100, Simon Willison 100, Zvi Mowshowitz 75, xlr8harder 75, Max Woolf 25) | 194 | 2026-04-23 |
| 16 | Composer 2.5 | Cursor | 82.3 | 81.8 | 83.7 | 93 | 46.8 | 95.6 | 71.3 (Every 75) | 97 | 2026-05-18 |
| 17 | Kimi K2.6 | Moonshot AI | 82.3 | 85.4 | 73.2 | 84.3 | 68.1 | 95.4 | 72.3 (Peter Gostev 75, Simon Willison 75) | 118 | 2026-04-22 |
| 18 | Gemma 4 | Google DeepMind | 82.3 | 85.3 | 73.1 | 93.8 | 69.4 | 81.1 | 71.1 (xlr8harder 100, Simon Willison 80, Ethan Mollick 75, Nathan Lambert 75, Max Woolf 75, Peter Gostev 25) | 196 | 2026-04-02 |
| 19 | GPT-6.1 Sol | OpenAI | 80.9 | 80.7 | 81.7 | 73.8 | 55.3 | 94.5 | 79.5 (Peter Gostev 100, Max Woolf 75) | 121 | 2026-09-29 |
| 20 | Kimi K3 | Moonshot AI | 80.3 | 81.6 | 76.4 | 90.7 | 63.4 | 91.4 | 64.4 (Nathan Lambert 100, Zvi Mowshowitz 75, Ethan Mollick 51, Simon Willison 25) | 204 | 2026-07-16 |
| 21 | Kimi K2 Thinking | Moonshot AI | 79.6 | 81.5 | 74.1 | 83.2 | 68.9 | 95.5 | 64.7 (Nathan Lambert 100, Zvi Mowshowitz 75, Sebastian Raschka 75, Simon Willison 42, Ethan Mollick 25) | 97 | 2025-11-06 |
| 22 | GPT-5.4 | OpenAI | 79 | 79 | 78.8 | 89.9 | 43.5 | 88.2 | 73.9 (Ethan Mollick 84, Nathan Lambert 75, Every 75, Max Woolf 75, Peter Gostev 68) | 136 | 2026-03-05 |
| 23 | GLM-5.3 | Z.ai | 78.4 | 80.8 | 71.3 | 76.3 | 83.8 | 89.1 | 71.3 (Nathan Lambert 75) | 149 | 2026-08-14 |
| 24 | Gemini 3.7 Flash | Google DeepMind | 77.6 | 76.7 | 80.3 | 81.6 | 44.5 | 95.9 | 71.3 (Simon Willison 75) | 143 | 2026-08-13 |
| 25 | Nemotron 3.5 Lightning | NVIDIA | 77.3 | 75.4 | 83 | 92.2 | 36.9 | 83.1 | 71.3 (Simon Willison 75) | 60 | 2026-08-11 |
| 26 | MiniMax-M2 | MiniMax | 77.3 | 77.3 | 77.1 | 83 |  |  | 72.3 (Simon Willison 75, Sebastian Raschka 75) | 54 | 2025-10-27 |
| 27 | GPT-5.2-Codex | OpenAI | 76.7 | 78.5 | 71.2 | 80.1 | 57.6 |  | 71.3 (Simon Willison 75) | 115 | 2025-12-18 |
| 28 | Composer 2 | Cursor | 76.1 | 71.5 | 89.9 | 77.6 | 47.3 | 77.6 | 71.3 (Every 75) | 71 | 2026-03-19 |
| 29 | Claude Opus 4.6 | Anthropic | 75.7 | 77.5 | 70.1 | 76.8 | 42 | 90.7 | 80.9 (Nathan Lambert 100, Ethan Mollick 90, Every 88, Zvi Mowshowitz 75, Simon Willison 71) | 209 | 2026-02-05 |
| 30 | DeepSeek V3.2 | DeepSeek | 74.8 | 78.6 | 63.3 | 72.6 | 67.7 |  | 71.3 (Zvi Mowshowitz 75) | 65 | 2025-12-01 |
| 31 | GLM-5.1 | Z.ai | 74.3 | 76.9 | 66.4 | 66.7 | 70.7 |  |  | 59 | 2026-03-27 |
| 32 | Muse Spark 1.1 | Meta | 74 | 74.8 | 71.4 | 77.8 | 48.6 |  | 71.3 (Simon Willison 75) | 88 | 2026-07-09 |
| 33 | DeepSeek V4 | DeepSeek | 74 | 73.4 | 75.7 | 83 | 82 | 89 | 50.9 (Simon Willison 75, Peter Gostev 0) | 153 | 2026-04-24 |
| 34 | GPT-6 Astra | OpenAI | 73.8 | 74.8 | 70.6 | 85.5 | 48.4 | 60.9 | 79.5 (Simon Willison 100, Every 75) | 211 | 2026-09-03 |
| 35 | Muse Spark 1.3 | Meta | 73.2 | 75 | 67.8 | 68.9 | 71.7 | 96.2 | 61.8 (Sam Paech 100, Simon Willison 75, xlr8harder 0) | 126 | 2026-09-02 |
| 36 | Gemini 3.1 Pro | Google DeepMind | 72.3 | 75.2 | 63.5 | 67.5 | 53 | 98.2 | 72.9 (Ethan Mollick 75, Simon Willison 75, Max Woolf 75) | 172 | 2026-02-19 |
| 37 | Nemotron 3 Super | NVIDIA | 71.6 | 73 | 67.2 | 85.8 |  |  | 61.3 (Nathan Lambert 75, Sebastian Raschka 36) | 59 | 2026-03-11 |
| 38 | Claude Haiku 5.5 | Anthropic | 71.4 | 70.2 | 75.3 | 84.9 | 53.3 | 79.2 | 58.1 (Max Woolf 75, Simon Willison 25) | 130 | 2026-10-07 |
| 39 | MiMo-V2.6-Pro | Xiaomi | 71.3 | 73.7 | 64 | 73.2 | 67.8 | 94.3 | 58.1 (Sebastian Raschka 75, xlr8harder 25) | 62 | 2026-09-21 |
| 40 | Nemotron 3 Ultra | NVIDIA | 70.7 | 72.4 | 65.6 | 70.4 |  |  | 72.3 (Sebastian Raschka 75, Nathan Lambert 75) | 54 | 2026-06-04 |
| 41 | Gemini Omni | Google DeepMind | 70.1 | 72.4 | 63 | 74.8 | 29.7 | 87.7 | 81.3 (Ethan Mollick 100) | 92 | 2026-05-19 |
| 42 | Fable 5.1 | Anthropic | 69.9 | 70.4 | 68.1 | 77.6 | 38.7 | 74.8 | 75.9 (Zvi Mowshowitz 89, Simon Willison 75, Every 75, Ethan Mollick 75) | 211 | 2026-09-01 |
| 43 | Fable 5 | Anthropic | 69.9 | 70.8 | 67 | 72.4 | 46.5 | 57.4 | 84.8 (Every 100, Ethan Mollick 100, Zvi Mowshowitz 98, Simon Willison 75, Nathan Lambert 75) | 168 | 2026-06-09 |
| 44 | Claude Opus 4.8 | Anthropic | 69.6 | 67.2 | 76.9 | 69.3 | 33.3 | 77.2 | 78 (Every 100, Peter Gostev 100, Ethan Mollick 87, Simon Willison 75, Zvi Mowshowitz 42) | 149 | 2026-05-28 |
| 45 | Mythos 5 | Anthropic | 69.5 | 72.8 | 59.5 | 92.2 | 23.6 |  | 71.3 (Every 75) | 41 | 2026-06-09 |
| 46 | GPT-5.1-Codex-Max | OpenAI | 69.3 | 72.9 | 58.5 | 57.8 | 54.7 |  | 79.5 (Peter Gostev 100, Ethan Mollick 75) | 90 | 2025-11-19 |
| 47 | GPT-6.1 Terra | OpenAI | 69.2 | 64.9 | 81.9 | 74.1 | 49 | 80.6 | 58.1 (Peter Gostev 75, Max Woolf 25) | 128 | 2026-10-26 |
| 48 | GPT-6 Luna | OpenAI | 69 | 71.5 | 61.2 | 56.5 | 75 | 88.1 | 68.8 (Simon Willison 100, Zvi Mowshowitz 75, Max Woolf 75, xlr8harder 25) | 108 | 2026-09-22 |
| 49 | GLM-5 | Z.ai | 68.5 | 70 | 63.9 | 79.4 | 58.6 | 94.3 | 51.3 (Peter Gostev 25) | 105 | 2026-02-11 |
| 50 | Nemotron 3.5 Super VL | NVIDIA | 68.2 | 70.4 | 61.7 | 68.8 |  | 93.8 |  | 49 | 2026-10-15 |
| 51 | Nemotron 3 Nano | NVIDIA | 68 | 66.9 | 71.5 | 82.6 |  |  | 51.3 (xlr8harder 25) | 53 | 2025-12-15 |
| 52 | Claude Sonnet 4.6 | Anthropic | 67.9 | 70.7 | 59.2 | 70.2 | 42.1 | 92.5 | 71.3 (Every 75) | 123 | 2026-02-17 |
| 53 | Qwen3.5 | Alibaba | 67.9 | 69.4 | 63.4 | 73.8 | 67.8 |  | 51.3 (Simon Willison 25) | 69 | 2026-02-16 |
| 54 | Kimi K3.1 Pro | Moonshot AI | 67.5 | 71.3 | 55.9 | 70.4 | 51.9 | 93.1 | 65.9 (Peter Gostev 100, Simon Willison 100, Max Woolf 75, xlr8harder 25, Nathan Lambert 25) | 134 | 2026-10-23 |
| 55 | GPT-5.6 Terra | OpenAI | 66.5 | 66.5 |  | 84.7 | 44.4 |  | 51.3 (xlr8harder 25) | 41 | 2026-06-26 |
| 56 | GPT-5.6 Luna | OpenAI | 66.2 | 66.2 |  | 83.9 | 44.4 |  | 51.3 (xlr8harder 25) | 39 | 2026-06-26 |
| 57 | MAI-Cyber-1-Flash | Microsoft | 65.5 | 65.5 |  | 76.9 | 30.8 |  |  | 40 | 2026-07-27 |
| 58 | Claude Sonnet 5.5 | Anthropic | 65.4 | 65.3 | 65.7 | 85.5 | 44.9 | 83.5 | 50.7 (Every 75, Peter Gostev 25, Simon Willison 25) | 139 | 2026-09-28 |
| 59 | Gemini 3.8 Flash | Google DeepMind | 65.3 | 67.9 | 57.5 | 53.9 | 58.6 | 74.3 | 77.5 (Simon Willison 93, Ethan Mollick 75) | 175 | 2026-09-02 |
| 60 | Claude Mythos Preview | Anthropic | 64.8 | 64.8 |  | 66.2 | 27.6 | 80 | 77.7 (Every 100, Zvi Mowshowitz 100, Ethan Mollick 47) | 131 | 2026-04-07 |
| 61 | Muse Spark | Meta | 64.7 | 63.5 | 68.6 | 72.5 | 39.2 |  | 58.1 (Simon Willison 75, Ethan Mollick 25) | 102 | 2026-04-08 |
| 62 | Claude Opus 5 | Anthropic | 63.9 | 68 | 51.6 | 70.3 | 41.9 | 92 | 66 (Nathan Lambert 100, Ethan Mollick 75, Zvi Mowshowitz 75, Peter Gostev 51, Every 25) | 181 | 2026-07-24 |
| 63 | GPT-Rosalind | OpenAI | 62.6 | 62.6 |  | 69 | 34.3 |  |  | 44 | 2026-04-19 |
| 64 | Gemini 4 Argon | Google DeepMind | 61.6 | 61.8 | 61.2 | 72.8 | 33.9 | 81.2 |  | 119 | 2026-10-01 |
| 65 | MiniMax M3.1 Flash | MiniMax | 61.1 | 58.6 | 68.7 | 75.9 | 16.1 |  |  | 81 | 2026-10-19 |
| 66 | GPT-5.2 | OpenAI | 61.1 | 60.4 | 62.9 | 53.7 | 52.1 | 75.4 | 65.8 (Simon Willison 88, xlr8harder 75, Every 75, Peter Gostev 62, Zvi Mowshowitz 25) | 146 | 2025-12-11 |
| 67 | MAI-Code-1-Flash | Microsoft | 61 | 61 |  | 77.7 | 19.4 |  |  | 66 | 2026-06-02 |
| 68 | Muse Spark 1.2 | Meta | 59.2 | 59 | 59.8 | 75.8 | 39.4 | 78 | 51.9 (Simon Willison 54, Max Woolf 25) | 107 | 2026-08-05 |
| 69 | Qwen3.8-Max | Alibaba | 56 | 58.9 | 47.5 | 51.2 |  |  |  | 46 | 2026-07-19 |
| 70 | MAI-Thinking-1 | Microsoft | 55.6 | 55.6 |  | 66.8 | 31.8 |  | 51.3 (Ethan Mollick 25) | 53 | 2026-08-12 |
| 71 | GPT-6 Sol | OpenAI | 55.4 | 59.4 | 43.4 | 40.8 | 51.4 | 77.7 | 73.3 (Every 100, Simon Willison 100, Zvi Mowshowitz 75, xlr8harder 25) | 122 | 2026-09-22 |
| 72 | Muse Spark Open | Meta | 52.7 | 52.7 |  | 79.5 | 33 | 75.5 | 41.3 (Nathan Lambert 0) | 113 | 2026-10-15 |
| 73 | GPT-5.1 | OpenAI | 50.9 | 48.5 | 58.2 | 64.4 | 20.6 | 88.9 | 45.7 (Ethan Mollick 75, Nathan Lambert 44, Simon Willison 25, xlr8harder 25, Max Woolf 25) | 148 | 2025-11-12 |
| 74 | Mistral Large 4 | Mistral AI | 49.4 | 49.9 | 48.1 | 33.1 | 47.7 | 85.8 | 58.1 (Simon Willison 75, Peter Gostev 25) | 116 | 2026-10-06 |
| 75 | GPT-5.6 Sol | OpenAI | 48.2 | 51.6 | 38 | 64.8 | 43.7 | 46.3 | 58.1 (Max Woolf 75, Zvi Mowshowitz 25) | 90 | 2026-06-26 |
| 76 | Inkling | Thinking Machines | 48.1 | 55.6 | 25.5 | 52.7 |  |  | 52.1 (Nathan Lambert 75, Sebastian Raschka 75, xlr8harder 25, Ethan Mollick 8) | 41 | 2026-07-15 |
| 77 | Grok 4.1 | SpaceXAI | 44.5 | 41.3 | 54.2 | 72.2 | 36.7 | 58.2 | 30.1 (Ethan Mollick 25, Max Woolf 7, Peter Gostev 0) | 95 | 2025-11-17 |
| 78 | Gemini 3.5 Flash | Google DeepMind | 43.4 | 44.7 | 39.4 | 41.8 | 26.1 | 92.9 | 50.6 (Ethan Mollick 75, Zvi Mowshowitz 75, Simon Willison 25, Peter Gostev 0) | 154 | 2026-05-19 |
| 79 | Claude Opus 4.7 | Anthropic | 43 | 45.2 | 36.3 | 50.9 | 22.7 | 80 | 52.9 (Every 75, Zvi Mowshowitz 75, Ethan Mollick 66, Simon Willison 25, xlr8harder 0) | 204 | 2026-04-17 |
| 80 | Gemini 3.6 Flash | Google DeepMind | 39.5 | 40.4 | 36.7 | 13.1 | 47.5 | 90.5 |  | 133 | 2026-07-21 |
| 81 | Muse Wonder | Meta | 37 | 42.5 | 20.6 | 26.1 | 38 | 82.6 | 56.3 (Simon Willison 75, Lech Mazur 75, Every 0) | 86 | 2026-10-12 |
| 82 | Grok 4.7 | SpaceXAI | 33.4 | 31.1 | 40.4 | 36.6 | 33.3 | 52 |  | 123 | 2026-09-21 |
| 83 | Claude Sonnet 5 | Anthropic | 25 | 24.9 | 25.3 | 22.5 | 21.2 | 76.9 | 36.6 (Zvi Mowshowitz 25, Every 0) | 170 | 2026-06-30 |

JSON with sources, verdicts and quotes: https://handyai.news/api/benchmarks/sentiment-bench
