HumainE Leaderboard: leaderboard
HumainE Leaderboard — human-centered evaluation of conversational systems across user success, satisfaction and preference criteria; overall score.
Metric: HumainE Score. Source: huggingface.co. Saturation forecast: Around 2034. 54 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 35.38 | #58 |
| 2 | Gemini 3.5 Flash | 34.81 | #60 |
| 3 | Gemini 3 Pro | 33.8 | #80 |
| 4 | Claude Fable 5 | 33.52 | #16 |
| 5 | Grok 4.20 | 33.32 | #175 |
| 6 | Gemini 2.5 Pro | 33.23 | #152 |
| 7 | Qwen 3.7 Max | 33.03 | #75 |
| 8 | Mistral Large 3 | 32.68 | #401 |
| 9 | DeepSeek V4 Flash | 32.39 | #137 |
| 10 | Qwen 3 235B A22B 2507 | 32.16 | #260 |
| 11 | GPT-5.2 Instant | 32.12 | #210 |
| 12 | Claude Opus 4.7 | 31.76 | #49 |
| 13 | DeepSeek V4 Pro | 31.05 | #102 |
| 14 | DeepSeek V3 (0324) | 30.95 | #340 |
| 15 | Magistral Medium | 30.88 | #494 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=humaine-leaderboard · How It Works · Data refreshed daily, snapshot 2026-10-20.