← All comparisons
Nova 2.0 Pro Preview (Non-reasoning) vs Claude 4.5 Sonnet (Reasoning)
Amazon vs Anthropic — side-by-side benchmark comparison
| Nova 2.0 Pro Preview (Non-reasoning) | Claude 4.5 Sonnet (Reasoning) | |
|---|---|---|
| Intelligence Index | 23.1 | 43.0 |
| Coding Index | 20.5 | 38.6 |
| Math Index | 30.7 | 88.0 |
| Output speed (tok/s) | 154.7 | 55.0 |
| Blended price ($/1M) | $3.44 | $6.56 |
| Time to first token (s) | 0.71s | 7.02s |
| aime | — | — |
| aime 25 | 30.7% | 88.0% |
| artificial analysis coding index | 20.50 | 38.60 |
| artificial analysis intelligence index | 23.10 | 43.00 |
| artificial analysis math index | 30.70 | 88.00 |
| gpqa | 63.6% | 83.4% |
| hle | 4.0% | 17.3% |
| ifbench | 52.0% | 57.3% |
| lcr | 28.3% | 65.7% |
| livecodebench | 47.3% | 71.4% |
| math 500 | — | — |
| mmlu pro | 77.2% | 87.5% |
| scicode | 28.1% | 44.7% |
| tau2 | 71.6% | 78.1% |
| terminalbench hard | 16.7% | 35.6% |
Benchmark data from Artificial Analysis.