← All comparisons
Devstral Small 2 vs Llama 3.1 Tulu3 405B
Mistral vs Allen Institute for AI — side-by-side benchmark comparison
| Devstral Small 2 | Llama 3.1 Tulu3 405B | |
|---|---|---|
| Intelligence Index | 19.5 | 14.1 |
| Coding Index | 20.7 | — |
| Math Index | 34.3 | — |
| Output speed (tok/s) | 75.2 | 0.0 |
| Blended price ($/1M) | $0.00 | $0.00 |
| Time to first token (s) | 0.73s | 0.00s |
| aime | — | 13.3% |
| aime 25 | 34.3% | — |
| artificial analysis coding index | 20.70 | — |
| artificial analysis intelligence index | 19.50 | 14.10 |
| artificial analysis math index | 34.30 | — |
| gpqa | 53.2% | 51.6% |
| hle | 3.4% | 3.5% |
| ifbench | 31.2% | — |
| lcr | 24.0% | — |
| livecodebench | 34.8% | 29.1% |
| math 500 | — | 77.8% |
| mmlu pro | 67.8% | 71.6% |
| scicode | 28.8% | 30.2% |
| tau2 | 23.4% | — |
| terminalbench hard | 16.7% | — |
Benchmark data from Artificial Analysis.