← All comparisons
gpt-oss-120b (high) vs Hermes 4 - Llama-3.1 405B (Reasoning)
OpenAI vs Nous Research — side-by-side benchmark comparison
| gpt-oss-120b (high) | Hermes 4 - Llama-3.1 405B (Reasoning) | |
|---|---|---|
| Intelligence Index | 33.3 | 18.6 |
| Coding Index | 28.6 | 16.0 |
| Math Index | 93.4 | 69.7 |
| Output speed (tok/s) | 356.8 | 38.6 |
| Blended price ($/1M) | $0.26 | $1.50 |
| Time to first token (s) | 0.51s | 0.79s |
| aime | — | — |
| aime 25 | 93.4% | 69.7% |
| artificial analysis coding index | 28.60 | 16.00 |
| artificial analysis intelligence index | 33.30 | 18.60 |
| artificial analysis math index | 93.40 | 69.70 |
| gpqa | 78.2% | 72.7% |
| hle | 18.5% | 10.3% |
| ifbench | 69.0% | 32.7% |
| lcr | 50.7% | 20.7% |
| livecodebench | 87.8% | 68.6% |
| math 500 | — | — |
| mmlu pro | 80.8% | 82.9% |
| scicode | 38.9% | 25.2% |
| tau2 | 65.8% | 22.2% |
| terminalbench hard | 23.5% | 11.4% |
Benchmark data from Artificial Analysis.