← All comparisons
Nova 2.0 Pro Preview (medium) vs Claude 4.5 Sonnet (Reasoning)
Amazon vs Anthropic — side-by-side benchmark comparison
| Nova 2.0 Pro Preview (medium) | Claude 4.5 Sonnet (Reasoning) | |
|---|---|---|
| Intelligence Index | 35.7 | 43.0 |
| Coding Index | 30.4 | 38.6 |
| Math Index | 89.0 | 88.0 |
| Output speed (tok/s) | 124.9 | 55.0 |
| Blended price ($/1M) | $3.44 | $6.56 |
| Time to first token (s) | 16.70s | 7.02s |
| aime | — | — |
| aime 25 | 89.0% | 88.0% |
| artificial analysis coding index | 30.40 | 38.60 |
| artificial analysis intelligence index | 35.70 | 43.00 |
| artificial analysis math index | 89.00 | 88.00 |
| gpqa | 78.5% | 83.4% |
| hle | 8.9% | 17.3% |
| ifbench | 79.0% | 57.3% |
| lcr | 54.3% | 65.7% |
| livecodebench | 73.0% | 71.4% |
| math 500 | — | — |
| mmlu pro | 83.0% | 87.5% |
| scicode | 42.7% | 44.7% |
| tau2 | 92.7% | 78.1% |
| terminalbench hard | 24.2% | 35.6% |
Benchmark data from Artificial Analysis.