HUMAINE: leaderboard
Prolific human-centered LLM benchmark measuring real-world user success, satisfaction, engagement, and preference patterns across demographically broad feedback.
Metric: Overall score. Source: huggingface.co. Source: benchmarklist.com. Saturation forecast: Around August 2027. 45 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 | 3.79 | #124 |
| 2 | DeepSeek R1 0528 | 3.79 | #222 |
| 3 | Gemini 2.5 Pro | 3.76 | #152 |
| 4 | Gemini 3.1 Pro (Preview) | 3.73 | #58 |
| 5 | Grok 4 | 3.72 | #181 |
| 6 | Kimi K2 | 3.71 | #243 |
| 7 | GPT-5.4 | 3.7 | #79 |
| 8 | GPT-5.5 | 3.7 | #31 |
| 9 | Magistral Medium | 3.69 | #494 |
| 10 | Claude Opus 4.7 | 3.68 | #49 |
| 11 | Grok 3 | 3.67 | #304 |
| 12 | Gemini 2.5 Flash | 3.66 | #245 |
| 13 | DeepSeek V4 Flash | 3.66 | #137 |
| 14 | DeepSeek V3 (0324) | 3.64 | #340 |
| 15 | GPT-5 Mini | 3.63 | #186 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=humaine · How It Works · Data refreshed daily, snapshot 2026-10-20.