← All comparisons
Claude 4.5 Haiku (Non-reasoning) vs Hermes 3 - Llama-3.1 70B
Anthropic vs Nous Research — side-by-side benchmark comparison
| Claude 4.5 Haiku (Non-reasoning) | Hermes 3 - Llama-3.1 70B | |
|---|---|---|
| Intelligence Index | 31.0 | 10.6 |
| Coding Index | 29.6 | — |
| Math Index | 39.0 | — |
| Output speed (tok/s) | 105.8 | 33.2 |
| Blended price ($/1M) | $2.19 | $0.30 |
| Time to first token (s) | 0.65s | 0.38s |
| aime | — | 2.3% |
| aime 25 | 39.0% | — |
| artificial analysis coding index | 29.60 | — |
| artificial analysis intelligence index | 31.00 | 10.60 |
| artificial analysis math index | 39.00 | — |
| gpqa | 64.6% | 40.1% |
| hle | 4.3% | 4.1% |
| ifbench | 42.0% | — |
| lcr | 43.7% | — |
| livecodebench | 51.1% | 18.8% |
| math 500 | — | 53.8% |
| mmlu pro | 80.0% | 57.1% |
| scicode | 34.4% | 23.1% |
| tau2 | 32.5% | — |
| terminalbench hard | 27.3% | — |
Benchmark data from Artificial Analysis.