Across 22 shared benchmarks, Claude 3.7 Sonnet scores higher on 5 and Qwen3.5 122B A10B on 16, with 1 level. The widest gap is HLE, where Qwen3.5 122B A10B scores 47.5 against 9.7. Qwen3.5 122B A10B is the cheaper of the two on tracked API pricing ($0.40 against $3.00 per million input tokens).
| Benchmark | Claude 3.7 Sonnet | Qwen3.5 122B A10B |
|---|---|---|
| AA Agentic Index | 37 | 21.3 |
| AA Intelligence | 27.6 | 32.8 |
| AA-LCR | 62.3 | 70.3 |
| AA-Omniscience | -0.7 | -41.5 |
| arena_vision | 1195 | 1246 |
| Artificial Analysis Coding Index | 36.4 | 45.7 |
| critpt | 0.9 | 0.9 |
| gdpval | 27.4 | 24.3 |
| GPQA Diamond | 84.8 | 86.6 |
| HLE | 9.7 | 47.5 |
| IFBench | 48.3 | 76.1 |
| ifeval | 93.2 | 93.4 |
| mmmlu | 86.1 | 86.7 |
| MMMU | 75 | 83.9 |
| MMMU-Pro | 60.1 | 76.9 |
| multichallenge | 51.6 | 61.5 |
| OmniScience Accuracy | 28.2 | 24.4 |
| OmniScience Non-Hallucination | 60 | 12.9 |
| scicode | 40.3 | 42 |
| SWE-bench Verified | 70.3 | 72 |
| Terminal-Bench Hard | 21.2 | 31.1 |
| τ²-Bench Telecom (AA run) | 54.7 | 93.6 |
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. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (direct, alibaba-official), otherwise the lowest tracked offer.