← All comparisons
Claude 4.5 Sonnet (Reasoning) vs DeepSeek R1 Distill Qwen 32B
Anthropic vs DeepSeek — side-by-side benchmark comparison
| Claude 4.5 Sonnet (Reasoning) | DeepSeek R1 Distill Qwen 32B | |
|---|---|---|
| Intelligence Index | 43.0 | 17.2 |
| Coding Index | 38.6 | — |
| Math Index | 88.0 | 63.0 |
| Output speed (tok/s) | 55.0 | 0.0 |
| Blended price ($/1M) | $6.56 | $0.00 |
| Time to first token (s) | 7.02s | 0.00s |
| aime | — | 68.7% |
| aime 25 | 88.0% | 63.0% |
| artificial analysis coding index | 38.60 | — |
| artificial analysis intelligence index | 43.00 | 17.20 |
| artificial analysis math index | 88.00 | 63.00 |
| gpqa | 83.4% | 61.5% |
| hle | 17.3% | 5.5% |
| ifbench | 57.3% | 22.9% |
| lcr | 65.7% | 9.7% |
| livecodebench | 71.4% | 27.0% |
| math 500 | — | 94.1% |
| mmlu pro | 87.5% | 73.9% |
| scicode | 44.7% | 37.6% |
| tau2 | 78.1% | — |
| terminalbench hard | 35.6% | — |
Benchmark data from Artificial Analysis.