| Benchmark | Claude 4 Opus | Qwen3.7 Max |
|---|---|---|
| AA Intelligence | 31.7 | 46.7 |
| AA-LCR | 40 | 74.7 |
| Artificial Analysis Coding Index | 34 | 66 |
| GPQA Diamond | 79.6 | 92.4 |
| HLE | 12.3 | 41.4 |
| IFBench | 53.7 | 80.5 |
| ifeval | 87.4 | 94.3 |
| livebench | 74.6 | 74.3 |
| LiveCodeBench v6 | 47.4 | 91.6 |
| longbench_v2 | 55.6 | 65.3 |
| mmlu_redux | 94.2 | 95 |
| MMLU-Pro | 86.6 | 89.6 |
| mmmlu | 88.8 | 90.3 |
| scicode | 40.9 | 53.5 |
| simplebench | 58.8 | 70.4 |
| supergpqa | 56.5 | 73.6 |
| SWE-bench Verified | 79.4 | 80.4 |
| Terminal-Bench Hard | 31.1 | 50.8 |
| τ²-Bench Telecom (AA run) | 73.4 | 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.