← All comparisons
DeepSeek V4 Pro (Non-reasoning) vs Claude 4.1 Opus (Non-reasoning)
DeepSeek vs Anthropic — side-by-side benchmark comparison
| DeepSeek V4 Pro (Non-reasoning) | Claude 4.1 Opus (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 39.3 | 36.0 |
| Coding Index | 38.4 | — |
| Math Index | — | — |
| Output speed (tok/s) | 53.3 | 44.7 |
| Blended price ($/1M) | $0.54 | $32.81 |
| Time to first token (s) | 1.18s | 1.63s |
| aime | — | — |
| aime 25 | — | — |
| artificial analysis coding index | 38.40 | — |
| artificial analysis intelligence index | 39.30 | 36.00 |
| artificial analysis math index | — | — |
| gpqa | 71.7% | — |
| hle | 7.7% | — |
| ifbench | 45.8% | — |
| lcr | 45.0% | — |
| livecodebench | — | — |
| math 500 | — | — |
| mmlu pro | — | — |
| scicode | 42.4% | — |
| tau2 | 91.2% | — |
| terminalbench hard | 36.4% | — |
Benchmark data from Artificial Analysis.