| Benchmark | Claude Opus 4.6 | Qwen2.5 Instruct 72B |
|---|---|---|
| AA Intelligence | 44.9 | 10 |
| AA-LCR | 74.3 | 20.3 |
| AA-Omniscience | 13.7 | -52.2 |
| Artificial Analysis Coding Index | 48.1 | 11.9 |
| Chinese SimpleQA (C-SimpleQA) | 76.4 | 52.2 |
| critpt | 12.6 | 0 |
| Fortress | 20.5 | 56.4 |
| GPQA Diamond | 91.3 | 49.1 |
| GSM8K | 0.6 | 95.8 |
| HLE | 53.1 | 4.2 |
| IFBench | 62.5 | 36.9 |
| livebench | 76.3 | 52.3 |
| LiveCodeBench | 88.8 | 55.5 |
| MMLU-Pro | 89.1 | 71.6 |
| mmmlu | 91.1 | 74.8 |
| OmniScience Accuracy | 47 | 17.6 |
| OmniScience Non-Hallucination | 37.2 | 15.3 |
| scicode | 52 | 26.7 |
| SWE-bench Verified | 80.8 | 23.8 |
| Terminal-Bench Hard | 65.4 | 4.5 |
| τ²-Bench Telecom (AA run) | 99.3 | 34.5 |
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.