← All comparisons
Claude 4.1 Opus (Non-reasoning) vs Claude 4 Sonnet (Reasoning)
Anthropic vs Anthropic — side-by-side benchmark comparison
| Claude 4.1 Opus (Non-reasoning) | Claude 4 Sonnet (Reasoning) | |
|---|---|---|
| Intelligence Index | 36.0 | 38.7 |
| Coding Index | — | 34.1 |
| Math Index | — | 74.3 |
| Output speed (tok/s) | 44.7 | 55.5 |
| Blended price ($/1M) | $32.81 | $6.56 |
| Time to first token (s) | 1.63s | 8.92s |
| aime | — | 77.3% |
| aime 25 | — | 74.3% |
| artificial analysis coding index | — | 34.10 |
| artificial analysis intelligence index | 36.00 | 38.70 |
| artificial analysis math index | — | 74.30 |
| gpqa | — | 77.7% |
| hle | — | 9.6% |
| ifbench | — | 54.7% |
| lcr | — | 64.7% |
| livecodebench | — | 65.5% |
| math 500 | — | 99.1% |
| mmlu pro | — | 84.2% |
| scicode | — | 40.0% |
| tau2 | — | 64.6% |
| terminalbench hard | — | 31.1% |
Benchmark data from Artificial Analysis.