AttuneBench (Omniscient): leaderboard
AttuneBench (2026) scores LLM emotional intelligence on genuine multi-turn conversations whose participants annotated their own emotions, preferred responses and judgments of how the model behaved. This board is the 0-100 Composite when the model also receives the pre-conversation mood and psychometric profile of the participant, on a 25-conversation subsample; higher is better.
Metric: Composite (0-100) = 100 x (0.24 emotion tracking + 0.49 turn-level verification + 0.27 holistic comprehension), the evaluated model stepping through real human-model conversations turn by turn against the participants' own annotations, Omniscient mode (the model also receives the participant's pre-conversation PANAS and psychometric profile), 25-conversation subsample; higher is better. Source: arxiv.org. Saturation forecast: Around 2031. 11 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.6 | 55.01 | #114 |
| 2 | Claude Opus 4.7 | 54.78 | #86 |
| 3 | Gemini 3.1 Pro (Preview) | 53.75 | #102 |
| 4 | Claude Haiku 4.5 | 53.38 | #587 |
| 5 | MiMo-V2-Pro | 53 | #356 |
| 6 | GPT-5.5 | 52.85 | #56 |
| 7 | Qwen 2.5 72B | 52.67 | #1189 |
| 8 | Mistral Large | 51.53 | #2144 |
| 9 | Grok 4 | 50.31 | #321 |
| 10 | Claude Sonnet 4.6 | 49.34 | #179 |
| 11 | GPT-5.4 | 49.06 | #138 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=attunebench-omniscient · How It Works · Data refreshed daily, snapshot 2026-11-22.