Epoch AI - Rli: leaderboard
Metric: Score. Source: epoch.ai. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Fable 5 | 16.1 |
| 2 | GPT-5.5 | 6.25 |
| 3 | Gemini 3.7 Flash | 5 |
| 4 | GPT-5.2 (Medium) | 2.5 |
| 5 | GPT-5.2 | 2.08 |
| 6 | Claude Sonnet 4.5 (Unknown) | 2.08 |
| 7 | GPT-5 | 1.67 |
| 8 | Gemini 3 Pro (Preview) | 1.25 |
| 9 | Gemini 2.5 Pro (Preview 06-05) | 0.83 |
Interactive version: theaggregate.ai/benchmark?slug=epoch-ai-rli · How It Works · Data refreshed daily, snapshot 2026-09-05.