← All comparisons
Hermes 4 - Llama-3.1 70B (Reasoning) vs Llama 3.3 Nemotron Super 49B v1 (Reasoning)
Nous Research vs NVIDIA — side-by-side benchmark comparison
| Hermes 4 - Llama-3.1 70B (Reasoning) | Llama 3.3 Nemotron Super 49B v1 (Reasoning) | |
|---|---|---|
| Intelligence Index | 16.0 | 18.5 |
| Coding Index | 14.4 | 9.4 |
| Math Index | 68.7 | 54.7 |
| Output speed (tok/s) | 92.8 | 0.0 |
| Blended price ($/1M) | $0.20 | $0.00 |
| Time to first token (s) | 0.64s | 0.00s |
| aime | — | 58.3% |
| aime 25 | 68.7% | 54.7% |
| artificial analysis coding index | 14.40 | 9.40 |
| artificial analysis intelligence index | 16.00 | 18.50 |
| artificial analysis math index | 68.70 | 54.70 |
| gpqa | 69.9% | 64.3% |
| hle | 7.9% | 6.5% |
| ifbench | 31.3% | 38.1% |
| lcr | 6.7% | 17.0% |
| livecodebench | 65.3% | 27.7% |
| math 500 | — | 95.9% |
| mmlu pro | 81.1% | 78.5% |
| scicode | 34.1% | 28.2% |
| tau2 | 22.5% | 26.9% |
| terminalbench hard | 4.5% | 0.0% |
Benchmark data from Artificial Analysis.