← All comparisons
Hermes 3 - Llama-3.1 70B vs Jamba 1.6 Large
Nous Research vs AI21 Labs — side-by-side benchmark comparison
| Hermes 3 - Llama-3.1 70B | Jamba 1.6 Large | |
|---|---|---|
| Intelligence Index | 10.6 | 10.6 |
| Coding Index | — | — |
| Math Index | — | — |
| Output speed (tok/s) | 33.2 | 54.9 |
| Blended price ($/1M) | $0.30 | $3.50 |
| Time to first token (s) | 0.38s | 0.87s |
| aime | 2.3% | 4.7% |
| aime 25 | — | — |
| artificial analysis coding index | — | — |
| artificial analysis intelligence index | 10.60 | 10.60 |
| artificial analysis math index | — | — |
| gpqa | 40.1% | 38.7% |
| hle | 4.1% | 4.0% |
| ifbench | — | — |
| lcr | — | — |
| livecodebench | 18.8% | 17.2% |
| math 500 | 53.8% | 58.0% |
| mmlu pro | 57.1% | 56.5% |
| scicode | 23.1% | 18.4% |
| tau2 | — | — |
| terminalbench hard | — | — |
Benchmark data from Artificial Analysis.