← All comparisons
NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) vs Qwen3 235B A22B 2507 Instruct
NVIDIA vs Alibaba — side-by-side benchmark comparison
| NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) | Qwen3 235B A22B 2507 Instruct | |
|---|---|---|
| Intelligence Index | 13.2 | 25.0 |
| Coding Index | 15.8 | 22.1 |
| Math Index | 13.3 | 71.7 |
| Output speed (tok/s) | 93.8 | 57.0 |
| Blended price ($/1M) | $0.09 | $0.36 |
| Time to first token (s) | 0.26s | 1.34s |
| aime | — | 71.7% |
| aime 25 | 13.3% | 71.7% |
| artificial analysis coding index | 15.80 | 22.10 |
| artificial analysis intelligence index | 13.20 | 25.00 |
| artificial analysis math index | 13.30 | 71.70 |
| gpqa | 39.9% | 75.3% |
| hle | 4.6% | 10.6% |
| ifbench | 37.5% | 46.1% |
| lcr | 6.7% | 31.2% |
| livecodebench | 36.0% | 52.4% |
| math 500 | — | 98.0% |
| mmlu pro | 57.9% | 82.8% |
| scicode | 23.0% | 36.0% |
| tau2 | 25.4% | 33.3% |
| terminalbench hard | 12.1% | 15.2% |
Benchmark data from Artificial Analysis.