EmoBench-M — leaderboard
Emotional intelligence benchmark for multimodal LLMs evaluating foundational emotion recognition, conversational emotion understanding, and socially complex emotion analysis.
Metric: Average Score. Source: github.com. Status: years away from saturation. 26 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 70.5 |
| 2 | GPT-5.2 | 66.5 |
| 3 | Gemini 2.0 Flash | 62.3 |
| 4 | Gemini 1.5 Flash | 61.3 |
| 5 | Gemini 2.0 Flash (Thinking) | 60.6 |
| 6 | Qwen 3 VL 8B Instruct | 60.1 |
| 7 | Qwen 2.5 VL 32B Instruct | 56.7 |
| 8 | Qwen2.5-Omni-7B | 53.7 |
| 9 | Qwen2-Audio-7B-Instruct | 53 |
| 10 | Qwen 2.5 VL 7B Instruct | 52.9 |
| 11 | InternVL2.5-78B | 52.4 |
Interactive version: theaggregate.ai/benchmark?slug=emobench-m · How the rankings work · Data refreshed daily, snapshot 2026-07-22.