← All comparisons
Nova 2.0 Omni (medium) vs Hermes 4 - Llama-3.1 405B (Non-reasoning)
Amazon vs Nous Research — side-by-side benchmark comparison
| Nova 2.0 Omni (medium) | Hermes 4 - Llama-3.1 405B (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 28.0 | 17.6 |
| Coding Index | 15.1 | 18.1 |
| Math Index | 89.7 | 15.3 |
| Output speed (tok/s) | 0.0 | 40.8 |
| Blended price ($/1M) | $0.85 | $1.50 |
| Time to first token (s) | 0.00s | 0.73s |
| aime | — | — |
| aime 25 | 89.7% | 15.3% |
| artificial analysis coding index | 15.10 | 18.10 |
| artificial analysis intelligence index | 28.00 | 17.60 |
| artificial analysis math index | 89.70 | 15.30 |
| gpqa | 76.0% | 53.6% |
| hle | 6.8% | 4.2% |
| ifbench | 66.2% | 34.8% |
| lcr | 53.7% | 20.0% |
| livecodebench | 66.0% | 54.6% |
| math 500 | — | — |
| mmlu pro | 80.9% | 72.9% |
| scicode | 36.2% | 34.6% |
| tau2 | 80.4% | 26.6% |
| terminalbench hard | 4.5% | 9.8% |
Benchmark data from Artificial Analysis.