← All comparisons
NVIDIA Nemotron 3 Super 120B A12B (Reasoning) vs MiniMax-M2.5
NVIDIA vs MiniMax — side-by-side benchmark comparison
| NVIDIA Nemotron 3 Super 120B A12B (Reasoning) | MiniMax-M2.5 | |
|---|---|---|
| Intelligence Index | 36.0 | 41.9 |
| Coding Index | 31.2 | 37.4 |
| Math Index | — | — |
| Output speed (tok/s) | 188.0 | 190.8 |
| Blended price ($/1M) | $0.41 | $0.53 |
| Time to first token (s) | 1.06s | 2.38s |
| aime | — | — |
| aime 25 | — | — |
| artificial analysis coding index | 31.20 | 37.40 |
| artificial analysis intelligence index | 36.00 | 41.90 |
| artificial analysis math index | — | — |
| gpqa | 80.0% | 84.8% |
| hle | 19.2% | 19.1% |
| ifbench | 71.5% | 71.6% |
| lcr | 60.0% | 66.0% |
| livecodebench | — | — |
| math 500 | — | — |
| mmlu pro | — | — |
| scicode | 36.0% | 42.6% |
| tau2 | 67.8% | 95.3% |
| terminalbench hard | 28.8% | 34.8% |
Benchmark data from Artificial Analysis.