← All comparisons
Hermes 4 - Llama-3.1 405B (Non-reasoning) vs GPT-4o (Aug '24)
Nous Research vs OpenAI — side-by-side benchmark comparison
| Hermes 4 - Llama-3.1 405B (Non-reasoning) | GPT-4o (Aug '24) | |
|---|---|---|
| Intelligence Index | 17.6 | 18.6 |
| Coding Index | 18.1 | 16.6 |
| Math Index | 15.3 | — |
| Output speed (tok/s) | 40.8 | 117.5 |
| Blended price ($/1M) | $1.50 | $4.38 |
| Time to first token (s) | 0.73s | 0.60s |
| aime | — | 11.7% |
| aime 25 | 15.3% | — |
| artificial analysis coding index | 18.10 | 16.60 |
| artificial analysis intelligence index | 17.60 | 18.60 |
| artificial analysis math index | 15.30 | — |
| gpqa | 53.6% | 52.1% |
| hle | 4.2% | 2.9% |
| ifbench | 34.8% | 36.0% |
| lcr | 20.0% | 35.0% |
| livecodebench | 54.6% | 31.7% |
| math 500 | — | 79.5% |
| mmlu pro | 72.9% | — |
| scicode | 34.6% | 33.1% |
| tau2 | 26.6% | 28.9% |
| terminalbench hard | 9.8% | 8.3% |
Benchmark data from Artificial Analysis.