AttuneBench (Verbose): 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 of emotion tracking, turn-level verification and holistic comprehension when the model also writes out its reasoning, on a 50-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, Verbose mode (the model also writes out its reasoning), 50-conversation subsample; higher is better. Source: arxiv.org. Saturation forecast: Around 2031. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Claude Opus 4.753.95#85
2Claude Opus 4.653.5#113
3GPT-5.553.44#54
4MiMo-V2-Pro52.71#353
5Qwen 2.5 72B51.48#1188
6Claude Haiku 4.551.47#583
7Gemini 3.1 Pro (Preview)51.21#99
8Claude Sonnet 4.649.73#173
9Grok 449.7#319
10Mistral Large49.66#2127
11GPT-5.449.1#135

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=attunebench-verbose · How It Works · Data refreshed daily, snapshot 2026-11-21.