| Benchmark | Domain | Score | vs best recorded | Setting | Run | Source |
|---|---|---|---|---|---|---|
| BBH | reasoning | 38.0% | 43% | — | — | External ↗ |
| GSM8K | math | 34.4% | 36% | — | — | External ↗ |
| MMLU | knowledge | 47.9% | 54% | — | — | External ↗ |
| PIQA | other | 81.9% | 92% | — | — | External ↗ |
| Winogrande | other | 71.0% | 80% | — | — | External ↗ |
| OpenBookQA | other | 52.0% | 59% | — | — | External ↗ |
| ARC AI2 | other | 50.6% | 53% | — | — | External ↗ |
| TriviaQA | other | 73.6% | 84% | — | — | External ↗ |
Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.
0 observations (OpenRouter listing). History accumulates with every ingest run; a single point means no change has been observed yet.