← All comparisons
Nova 2.0 Pro Preview (low) vs DeepSeek R1 Distill Qwen 14B
Amazon vs DeepSeek — side-by-side benchmark comparison
| Nova 2.0 Pro Preview (low) | DeepSeek R1 Distill Qwen 14B | |
|---|---|---|
| Intelligence Index | 31.9 | 15.8 |
| Coding Index | 24.5 | — |
| Math Index | 63.3 | 55.7 |
| Output speed (tok/s) | 129.1 | 0.0 |
| Blended price ($/1M) | $3.44 | $0.00 |
| Time to first token (s) | 3.95s | 0.00s |
| aime | — | 66.7% |
| aime 25 | 63.3% | 55.7% |
| artificial analysis coding index | 24.50 | — |
| artificial analysis intelligence index | 31.90 | 15.80 |
| artificial analysis math index | 63.30 | 55.70 |
| gpqa | 75.1% | 48.4% |
| hle | 5.2% | 4.4% |
| ifbench | 79.6% | 22.1% |
| lcr | 61.7% | 7.0% |
| livecodebench | 63.8% | 37.6% |
| math 500 | — | 94.9% |
| mmlu pro | 82.2% | 74.0% |
| scicode | 38.7% | 23.9% |
| tau2 | 90.6% | — |
| terminalbench hard | 17.4% | — |
Benchmark data from Artificial Analysis.