| Benchmark | o1 | Qwen3.7 Max |
|---|---|---|
| AA Agentic Index | 31.1 | 30.9 |
| AA Intelligence | 24 | 46.7 |
| AA-LCR | 63.3 | 74.7 |
| AA-Omniscience | -10.6 | 13.5 |
| Artificial Analysis Coding Index | 39.7 | 66 |
| critpt | 0.3 | 13.4 |
| gdpval | 11.5 | 38.5 |
| GPQA Diamond | 75.7 | 92.4 |
| HLE | 7.7 | 41.4 |
| IFBench | 70.3 | 80.5 |
| livebench | 52.3 | 74.3 |
| mmmlu | 87.7 | 90.3 |
| OmniScience Accuracy | 34.7 | 31.1 |
| OmniScience Non-Hallucination | 30.7 | 74.4 |
| scicode | 35.8 | 53.5 |
| simplebench | 41.7 | 70.4 |
| SWE-bench Verified | 48.9 | 80.4 |
| Terminal-Bench Hard | 12.9 | 50.8 |
| τ²-Bench Telecom (AA run) | 62.6 | 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.