| Benchmark | o1 | Qwen3.7 Plus Preview |
|---|---|---|
| AA Agentic Index | 31.1 | 20.8 |
| AA Intelligence | 24 | 39.4 |
| AA-LCR | 63.3 | 69 |
| AA-Omniscience | -10.6 | 2.4 |
| arena_vision | 1193 | 1263 |
| Artificial Analysis Coding Index | 39.7 | 55.9 |
| critpt | 0.3 | 9.1 |
| gdpval | 11.5 | 22.2 |
| GPQA Diamond | 75.7 | 90.3 |
| HLE | 7.7 | 35.6 |
| IFBench | 70.3 | 79.1 |
| mmmlu | 87.7 | 89 |
| OmniScience Accuracy | 34.7 | 22.5 |
| OmniScience Non-Hallucination | 30.7 | 74.5 |
| scicode | 35.8 | 51.3 |
| SWE-bench Verified | 48.9 | 77.7 |
| Terminal-Bench Hard | 12.9 | 47 |
| τ²-Bench Telecom (AA run) | 62.6 | 93 |
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.