| Benchmark | Claude Sonnet 5 | Gemini 1.5 Pro |
|---|---|---|
| AA Intelligence | 55.3 | 10 |
| Artificial Analysis Coding Index | 71.5 | 23.6 |
| GPQA Diamond | 91.1 | 59.1 |
| HLE | 57.4 | 4.9 |
| MMMU-Pro | 77.3 | 55 |
| scicode | 53.6 | 29.5 |
| simplebench | 60.6 | 27.1 |
| SWE-bench Verified | 85.2 | 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.