| Benchmark | Gemini 1.5 Pro | Qwen3.7 Plus Preview |
|---|---|---|
| AA Intelligence | 10 | 39.4 |
| arena_vision | 1179 | 1263 |
| Artificial Analysis Coding Index | 23.6 | 55.9 |
| GPQA Diamond | 59.1 | 90.3 |
| HLE | 4.9 | 35.6 |
| ifeval | 89.4 | 94.6 |
| MMLU-Pro | 75.8 | 88.5 |
| MMMU-Pro | 55 | 80.5 |
| ocrbench_v2 | 51.6 | 67.1 |
| scicode | 29.5 | 51.3 |
| SWE-bench Verified | 34.2 | 77.7 |
| Video-MME | 78.6 | 88 |
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.