← All comparisons
Qwen3.5 397B A17B (Non-reasoning) vs Claude Opus 4.6 (Adaptive Reasoning, Max Effort)
Alibaba vs Anthropic — side-by-side benchmark comparison
| Qwen3.5 397B A17B (Non-reasoning) | Claude Opus 4.6 (Adaptive Reasoning, Max Effort) | |
|---|---|---|
| Intelligence Index | 40.1 | 52.9 |
| Coding Index | 37.4 | 48.1 |
| Math Index | — | — |
| Output speed (tok/s) | 53.5 | 54.8 |
| Blended price ($/1M) | $1.35 | $10.94 |
| Time to first token (s) | 1.85s | 11.69s |
| aime | — | — |
| aime 25 | — | — |
| artificial analysis coding index | 37.40 | 48.10 |
| artificial analysis intelligence index | 40.10 | 52.90 |
| artificial analysis math index | — | — |
| gpqa | 86.1% | 89.6% |
| hle | 18.8% | 36.7% |
| ifbench | 51.6% | 53.1% |
| lcr | 58.0% | 70.7% |
| livecodebench | — | — |
| math 500 | — | — |
| mmlu pro | — | — |
| scicode | 41.1% | 51.9% |
| tau2 | 83.9% | 92.1% |
| terminalbench hard | 35.6% | 46.2% |
Benchmark data from Artificial Analysis.