| Benchmark | Gemini 1.5 Pro | Qwen3.6 27B |
|---|---|---|
| AA Intelligence | 10 | 37.7 |
| Artificial Analysis Coding Index | 23.6 | 53.7 |
| GPQA Diamond | 59.1 | 87.8 |
| hellaswag | 93.3 | 90 |
| HLE | 4.9 | 24 |
| humaneval | 86.6 | 78.2 |
| MathVista | 70.6 | 87.4 |
| MMLU-Pro | 75.8 | 86.2 |
| MMMU | 68.4 | 82.9 |
| MMMU-Pro | 55 | 75.8 |
| OCRBench | 800 | 89.4 |
| scicode | 29.5 | 39.8 |
| simpleqa | 23.4 | 95.7 |
| SWE-bench Verified | 34.2 | 77.2 |
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.