| Benchmark | Claude Opus 4.6 | Gemini 1.5 Pro |
|---|---|---|
| AA Intelligence | 44.9 | 10 |
| ARC-AGI-2 | 68.8 | 0.8 |
| arena_vision | 1300 | 1179 |
| Artificial Analysis Coding Index | 48.1 | 23.6 |
| Chinese SimpleQA (C-SimpleQA) | 76.4 | 59.4 |
| Fortress | 20.5 | 53.9 |
| GPQA Diamond | 91.3 | 59.1 |
| GSM8K | 0.6 | 95.2 |
| HLE | 53.1 | 4.9 |
| LiveCodeBench | 88.8 | 30.5 |
| MMLU-Pro | 89.1 | 75.8 |
| MMMU-Pro | 77.3 | 55 |
| ocrbench_v2 | 48.4 | 51.6 |
| scicode | 52 | 29.5 |
| SEAL VISTA | 46.1 | 37.1 |
| simplebench | 67.6 | 27.1 |
| SWE-bench Verified | 80.8 | 34.2 |
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.