| Benchmark | Gemini 1.5 Pro | GPT-5.6 Sol |
|---|---|---|
| AA Intelligence | 10 | 61 |
| ARC-AGI-2 | 0.8 | 92.5 |
| Artificial Analysis Coding Index | 23.6 | 78.3 |
| GPQA Diamond | 59.1 | 94.6 |
| HLE | 4.9 | 49.5 |
| MMMU-Pro | 55 | 84.6 |
| scicode | 29.5 | 56.9 |
| simplebench | 27.1 | 64.8 |
| Video-MME | 78.6 | 89.5 |
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.