| Benchmark | Gemini 1.5 Pro | Kimi K2.6 |
|---|---|---|
| AA Intelligence | 10 | 45.1 |
| arena_vision | 1179 | 1264 |
| Artificial Analysis Coding Index | 23.6 | 61.8 |
| Chinese SimpleQA (C-SimpleQA) | 59.4 | 75.9 |
| GPQA Diamond | 59.1 | 91.1 |
| HLE | 4.9 | 37.5 |
| LiveCodeBench | 30.5 | 89.6 |
| MMLU-Pro | 75.8 | 87.1 |
| MMMU-Pro | 55 | 80.1 |
| scicode | 29.5 | 53.5 |
| SWE-bench Verified | 34.2 | 80.2 |
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.