← All comparisons
Magistral Small 1.2 vs Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)
Mistral vs NVIDIA — side-by-side benchmark comparison
| Magistral Small 1.2 | Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) | |
|---|---|---|
| Intelligence Index | 18.2 | 15.0 |
| Coding Index | 14.8 | 13.1 |
| Math Index | 80.3 | 63.7 |
| Output speed (tok/s) | 111.1 | 52.3 |
| Blended price ($/1M) | $0.75 | $0.90 |
| Time to first token (s) | 0.38s | 0.76s |
| aime | — | 74.7% |
| aime 25 | 80.3% | 63.7% |
| artificial analysis coding index | 14.80 | 13.10 |
| artificial analysis intelligence index | 18.20 | 15.00 |
| artificial analysis math index | 80.30 | 63.70 |
| gpqa | 66.3% | 72.8% |
| hle | 6.1% | 8.1% |
| ifbench | 44.4% | 38.2% |
| lcr | 16.3% | 7.3% |
| livecodebench | 72.3% | 64.1% |
| math 500 | — | 95.2% |
| mmlu pro | 76.8% | 82.5% |
| scicode | 35.2% | 34.7% |
| tau2 | 27.8% | 11.4% |
| terminalbench hard | 4.5% | 2.3% |
Benchmark data from Artificial Analysis.