| Benchmark | o4-mini | Qwen3.7 Max |
|---|---|---|
| AA Agentic Index | 36.1 | 30.9 |
| AA Intelligence | 26.1 | 46.7 |
| AA-LCR | 60 | 74.7 |
| AA-Omniscience | -35.7 | 13.5 |
| Artificial Analysis Coding Index | 25.6 | 66 |
| critpt | 0.6 | 13.4 |
| gdpval | 25.4 | 38.5 |
| GPQA Diamond | 81.4 | 92.4 |
| HLE | 16.5 | 41.4 |
| IFBench | 68.7 | 80.5 |
| mmlu_prox | 69.3 | 87 |
| OmniScience Accuracy | 24.8 | 31.1 |
| OmniScience Non-Hallucination | 19.5 | 74.4 |
| scicode | 46.5 | 53.5 |
| simplebench | 38.7 | 70.4 |
| simpleqa_verified | 23.9 | 58.5 |
| SWE-bench Verified | 68.1 | 80.4 |
| Terminal-Bench Hard | 15.2 | 50.8 |
| τ²-Bench Telecom (AA run) | 55.6 | 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.