| Benchmark | Gemini 1.5 Pro | GPT-5.1 |
|---|---|---|
| AA Intelligence | 10 | 37.5 |
| ARC-AGI-2 | 0.8 | 17.6 |
| arena_vision | 1179 | 1250 |
| Artificial Analysis Coding Index | 23.6 | 49.4 |
| Fortress | 53.9 | 25.7 |
| GPQA Diamond | 59.1 | 88.1 |
| HLE | 4.9 | 28.5 |
| MMLU-Pro | 75.8 | 87 |
| MMMU | 68.4 | 85.4 |
| MMMU-Pro | 55 | 75.5 |
| scicode | 29.5 | 43.3 |
| SEAL VISTA | 37.1 | 43.8 |
| simplebench | 27.1 | 53.2 |
| SWE-bench Verified | 34.2 | 76.3 |
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.