Epoch AI - ECI — leaderboard
Epoch Capabilities Index: a composite metric using Item Response Theory across 37 benchmarks to produce a single general capability score - like an IQ test for AI models.
Metric: ECI Score. Source: epoch.ai. Status: saturated. 426 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.6 Sol | 162.06 |
| 2 | GPT-5.6 Sol (xHigh) | 162.06 |
| 3 | GPT-5.6 Sol (Max) | 162.06 |
| 4 | GPT-5.6 Sol (High) | 162.06 |
| 5 | GPT-5.6 Sol (Medium) | 162.06 |
| 6 | GPT-5.6 Sol (Low) | 162.06 |
| 7 | GPT-5.5 Pro | 160.93 |
| 8 | GPT-5.5 Pro (xHigh) | 160.93 |
| 9 | Claude Fable 5 | 160.75 |
| 10 | Claude Fable 5 (Max) | 160.75 |
| 11 | Claude Fable 5 (High) | 160.75 |
| 12 | GPT-5.5 | 158.52 |
| 13 | GPT-5.5 (xHigh) | 158.52 |
| 14 | GPT-5.5 (High) | 158.52 |
| 15 | GPT-5.5 (Medium) | 158.52 |
Interactive version: theaggregate.ai/benchmark?slug=epoch-ai-eci · How the rankings work · Data refreshed daily, snapshot 2026-07-22.