← All comparisons
Hermes 4 - Llama-3.1 405B (Reasoning) vs GLM-4.6V (Non-reasoning)
Nous Research vs Z AI — side-by-side benchmark comparison
| Hermes 4 - Llama-3.1 405B (Reasoning) | GLM-4.6V (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 18.6 | 17.1 |
| Coding Index | 16.0 | 11.1 |
| Math Index | 69.7 | 26.3 |
| Output speed (tok/s) | 38.6 | 38.5 |
| Blended price ($/1M) | $1.50 | $0.45 |
| Time to first token (s) | 0.79s | 1.39s |
| aime | — | — |
| aime 25 | 69.7% | 26.3% |
| artificial analysis coding index | 16.00 | 11.10 |
| artificial analysis intelligence index | 18.60 | 17.10 |
| artificial analysis math index | 69.70 | 26.30 |
| gpqa | 72.7% | 56.6% |
| hle | 10.3% | 3.7% |
| ifbench | 32.7% | 27.9% |
| lcr | 20.7% | 12.3% |
| livecodebench | 68.6% | 41.1% |
| math 500 | — | — |
| mmlu pro | 82.9% | 75.2% |
| scicode | 25.2% | 27.2% |
| tau2 | 22.2% | 30.7% |
| terminalbench hard | 11.4% | 3.0% |
Benchmark data from Artificial Analysis.