← All comparisons
Claude 4.5 Sonnet (Reasoning) vs Claude 4 Sonnet (Reasoning)
Anthropic vs Anthropic — side-by-side benchmark comparison
| Claude 4.5 Sonnet (Reasoning) | Claude 4 Sonnet (Reasoning) | |
|---|---|---|
| Intelligence Index | 43.0 | 38.7 |
| Coding Index | 38.6 | 34.1 |
| Math Index | 88.0 | 74.3 |
| Output speed (tok/s) | 55.0 | 55.5 |
| Blended price ($/1M) | $6.56 | $6.56 |
| Time to first token (s) | 7.02s | 8.92s |
| aime | — | 77.3% |
| aime 25 | 88.0% | 74.3% |
| artificial analysis coding index | 38.60 | 34.10 |
| artificial analysis intelligence index | 43.00 | 38.70 |
| artificial analysis math index | 88.00 | 74.30 |
| gpqa | 83.4% | 77.7% |
| hle | 17.3% | 9.6% |
| ifbench | 57.3% | 54.7% |
| lcr | 65.7% | 64.7% |
| livecodebench | 71.4% | 65.5% |
| math 500 | — | 99.1% |
| mmlu pro | 87.5% | 84.2% |
| scicode | 44.7% | 40.0% |
| tau2 | 78.1% | 64.6% |
| terminalbench hard | 35.6% | 31.1% |
Benchmark data from Artificial Analysis.