Benchmarks
RLI
Models scored
10
Best score
0.16
Metric
Score
RLI is an AI evaluation tracked by Epoch AI, with 10 scored model versions on record. The scores here are reported by third parties rather than produced by Epoch AI running the evaluation itself.
Scored models span June 2025 to June 2026. The best score recorded is 0.16, measured as Score. Models come from Anthropic, OpenAI, Google DeepMind.
Full record
- Score metric
- Score
- Models scored
- 10
- Best score recorded
- 0.16
- Earliest model scored
- Jun 5, 2025
- Most recent model scored
- Jun 9, 2026
- Third-party reported
- Yes
Leaderboard
| 01 | claude-fable-5 | Anthropic | 0.16 |
| 02 | claude-opus-4-8_unknown | Anthropic | 0.08 |
| 03 | claude-opus-4-6_unknown | Anthropic | 0.04 |
| 04 | claude-opus-4-5-20251101_unknown | Anthropic | 0.04 |
| 05 | gpt-5.2-2025-12-11_medium | OpenAI | 0.03 |
| 06 | gpt-5.2-2025-12-11_unknown | OpenAI | 0.02 |
| 07 | claude-sonnet-4-5-20250929_unknown | Anthropic | 0.02 |
| 08 | gpt-5-2025-08-07_unknown | OpenAI | 0.02 |
| 09 | gemini-3-pro-preview | Google DeepMind | 0.01 |
| 10 | gemini-2.5-pro-preview-06-05 | Google DeepMind | 0.01 |
More benchmarks
SourceEpoch AI, 'AI Benchmarking Hub'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/benchmarks. Licensed under CC BY 4.0.
← All benchmarks