| Benchmark | Claude Opus 4.8 | o1 |
|---|---|---|
| AA Agentic Index | 49.4 | 31.1 |
| AA Intelligence | 57.3 | 24 |
| AA-LCR | 73 | 63.3 |
| AA-Omniscience | 28.8 | -10.6 |
| arena_vision | 1280 | 1193 |
| Artificial Analysis Coding Index | 74.3 | 39.7 |
| critpt | 20.9 | 0.3 |
| Fortress | 18.2 | 19.4 |
| gdpval | 54.2 | 11.5 |
| GPQA Diamond | 93.6 | 75.7 |
| HLE | 57.9 | 7.7 |
| IFBench | 62.2 | 70.3 |
| livebench | 77.2 | 52.3 |
| OmniScience Accuracy | 48.8 | 34.7 |
| OmniScience Non-Hallucination | 60.7 | 30.7 |
| scicode | 53.5 | 35.8 |
| simplebench | 64.8 | 41.7 |
| SWE-bench Verified | 88.6 | 48.9 |
| Terminal-Bench Hard | 58.3 | 12.9 |
| τ²-Bench Telecom (AA run) | 94.4 | 62.6 |
Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.