← All comparisons
DeepHermes 3 - Mistral 24B Preview (Non-reasoning) vs Claude 3.5 Sonnet (June '24)
Nous Research vs Anthropic — side-by-side benchmark comparison
| DeepHermes 3 - Mistral 24B Preview (Non-reasoning) | Claude 3.5 Sonnet (June '24) | |
|---|---|---|
| Intelligence Index | 10.9 | 14.2 |
| Coding Index | — | 26.0 |
| Math Index | — | — |
| Output speed (tok/s) | 0.0 | 0.0 |
| Blended price ($/1M) | $0.00 | $6.56 |
| Time to first token (s) | 0.00s | 0.00s |
| aime | 4.7% | 9.7% |
| aime 25 | — | — |
| artificial analysis coding index | — | 26.00 |
| artificial analysis intelligence index | 10.90 | 14.20 |
| artificial analysis math index | — | — |
| gpqa | 38.2% | 56.0% |
| hle | 3.9% | 3.7% |
| ifbench | — | — |
| lcr | — | — |
| livecodebench | 19.5% | — |
| math 500 | 59.5% | 69.5% |
| mmlu pro | 58.0% | 75.1% |
| scicode | 22.8% | 31.6% |
| tau2 | — | — |
| terminalbench hard | — | — |
Benchmark data from Artificial Analysis.