← All comparisons
Grok 4.3 (medium) vs DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)
xAI vs Nous Research — side-by-side benchmark comparison
| Grok 4.3 (medium) | DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 48.8 | 7.6 |
| Coding Index | 35.1 | — |
| Math Index | — | — |
| Output speed (tok/s) | 112.5 | 0.0 |
| Blended price ($/1M) | $1.56 | $0.00 |
| Time to first token (s) | 17.68s | 0.00s |
| aime | — | 0.0% |
| aime 25 | — | — |
| artificial analysis coding index | 35.10 | — |
| artificial analysis intelligence index | 48.80 | 7.60 |
| artificial analysis math index | — | — |
| gpqa | 89.0% | 27.0% |
| hle | 28.1% | 4.3% |
| ifbench | 83.3% | — |
| lcr | 65.0% | — |
| livecodebench | — | 8.5% |
| math 500 | — | 21.8% |
| mmlu pro | — | 36.5% |
| scicode | 44.6% | 9.1% |
| tau2 | 91.2% | — |
| terminalbench hard | 30.3% | — |
Benchmark data from Artificial Analysis.