| Benchmark | Gemini 2.5 Flash | Qwen3.7 Max |
|---|---|---|
| AA Agentic Index | 18.8 | 30.9 |
| AA Intelligence | 17.7 | 46.7 |
| AA-LCR | 65.7 | 74.7 |
| AA-Omniscience | -29.8 | 13.5 |
| arena_text_factuality | 1437 | 1479 |
| Artificial Analysis Coding Index | 22.2 | 66 |
| critpt | 1.4 | 13.4 |
| gdpval | 11.9 | 38.5 |
| GPQA Diamond | 82.8 | 92.4 |
| HLE | 12.1 | 41.4 |
| IFBench | 50.3 | 80.5 |
| OmniScience Accuracy | 26.1 | 31.1 |
| OmniScience Non-Hallucination | 24.7 | 74.4 |
| scicode | 35.9 | 53.5 |
| simplebench | 41.2 | 70.4 |
| SWE-bench Verified | 60.4 | 80.4 |
| Terminal-Bench Hard | 13.6 | 50.8 |
| τ²-Bench Telecom (AA run) | 31.6 | 94.7 |
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.