EQ-Bench — leaderboard
Emotional intelligence benchmark testing how well models understand and process complex emotional scenarios and nuanced human interactions.
Metric: Normalized Elo (self-reported). Source: benchmarklist.com. Status: saturation imminent. 77 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Fable 5 | 2069.4 |
| 2 | Claude Opus 4.8 | 2043.8 |
| 3 | Claude Opus 4.7 | 1911.5 |
| 4 | Claude Opus 4.6 | 1744.2 |
| 5 | Claude Sonnet 4.6 | 1736.6 |
| 6 | DeepSeek V4 Pro | 1599.3 |
| 7 | GPT-5.5 | 1590.6 |
| 8 | GPT-5.4 | 1586.7 |
| 9 | GPT-5.2 | 1567.8 |
| 10 | GLM-5.1 | 1560.7 |
| 11 | GPT-5.1 | 1553.8 |
| 12 | Claude Opus 4.5 | 1543.3 |
| 13 | Gemini 2.5 Pro (Preview 06-05) | 1540.8 |
| 14 | Gemini 3.1 Pro (Preview) | 1535.9 |
| 15 | GLM-5 | 1532.7 |
Interactive version: theaggregate.ai/benchmark?slug=eq-bench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.