← All comparisons
Hermes 4 - Llama-3.1 70B (Reasoning) vs Qwen3 VL 235B A22B (Reasoning)
Nous Research vs Alibaba — side-by-side benchmark comparison
| Hermes 4 - Llama-3.1 70B (Reasoning) | Qwen3 VL 235B A22B (Reasoning) | |
|---|---|---|
| Intelligence Index | 16.0 | 27.6 |
| Coding Index | 14.4 | 20.9 |
| Math Index | 68.7 | 88.3 |
| Output speed (tok/s) | 92.8 | 35.6 |
| Blended price ($/1M) | $0.20 | $2.17 |
| Time to first token (s) | 0.64s | 5.14s |
| aime | — | — |
| aime 25 | 68.7% | 88.3% |
| artificial analysis coding index | 14.40 | 20.90 |
| artificial analysis intelligence index | 16.00 | 27.60 |
| artificial analysis math index | 68.70 | 88.30 |
| gpqa | 69.9% | 77.2% |
| hle | 7.9% | 10.1% |
| ifbench | 31.3% | 56.5% |
| lcr | 6.7% | 58.7% |
| livecodebench | 65.3% | 64.6% |
| math 500 | — | — |
| mmlu pro | 81.1% | 83.6% |
| scicode | 34.1% | 39.9% |
| tau2 | 22.5% | 54.1% |
| terminalbench hard | 4.5% | 11.4% |
Benchmark data from Artificial Analysis.