← All comparisons
Hermes 3 - Llama-3.1 70B vs Qwen3.5 27B (Non-reasoning)
Nous Research vs Alibaba — side-by-side benchmark comparison
| Hermes 3 - Llama-3.1 70B | Qwen3.5 27B (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 10.6 | 37.2 |
| Coding Index | — | 33.4 |
| Math Index | — | — |
| Output speed (tok/s) | 33.2 | 95.3 |
| Blended price ($/1M) | $0.30 | $0.88 |
| Time to first token (s) | 0.38s | 1.40s |
| aime | 2.3% | — |
| aime 25 | — | — |
| artificial analysis coding index | — | 33.40 |
| artificial analysis intelligence index | 10.60 | 37.20 |
| artificial analysis math index | — | — |
| gpqa | 40.1% | 84.2% |
| hle | 4.1% | 13.2% |
| ifbench | — | 46.9% |
| lcr | — | 55.7% |
| livecodebench | 18.8% | — |
| math 500 | 53.8% | — |
| mmlu pro | 57.1% | — |
| scicode | 23.1% | 36.7% |
| tau2 | — | 87.1% |
| terminalbench hard | — | 31.8% |
Benchmark data from Artificial Analysis.