← All comparisons
gpt-oss-20B (low) vs NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)
OpenAI vs NVIDIA — side-by-side benchmark comparison
| gpt-oss-20B (low) | NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 20.8 | 10.1 |
| Coding Index | 14.4 | 5.9 |
| Math Index | 62.3 | 26.7 |
| Output speed (tok/s) | 273.0 | 227.4 |
| Blended price ($/1M) | $0.10 | $0.30 |
| Time to first token (s) | 0.50s | 0.63s |
| aime | — | — |
| aime 25 | 62.3% | 26.7% |
| artificial analysis coding index | 14.40 | 5.90 |
| artificial analysis intelligence index | 20.80 | 10.10 |
| artificial analysis math index | 62.30 | 26.70 |
| gpqa | 61.1% | 43.9% |
| hle | 5.1% | 4.5% |
| ifbench | 57.8% | 25.9% |
| lcr | 31.0% | 17.0% |
| livecodebench | 65.2% | 34.5% |
| math 500 | — | — |
| mmlu pro | 71.8% | 64.9% |
| scicode | 34.0% | 17.6% |
| tau2 | 50.3% | 19.3% |
| terminalbench hard | 4.5% | 0.0% |
Benchmark data from Artificial Analysis.