← All comparisons
Qwen3 Omni 30B A3B Instruct vs Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)
Alibaba vs NVIDIA — side-by-side benchmark comparison
| Qwen3 Omni 30B A3B Instruct | Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning) | |
|---|---|---|
| Intelligence Index | 10.7 | 14.4 |
| Coding Index | 7.2 | — |
| Math Index | 52.3 | 50.0 |
| Output speed (tok/s) | 108.9 | 0.0 |
| Blended price ($/1M) | $0.43 | $0.00 |
| Time to first token (s) | 1.04s | 0.00s |
| aime | — | 70.7% |
| aime 25 | 52.3% | 50.0% |
| artificial analysis coding index | 7.20 | — |
| artificial analysis intelligence index | 10.70 | 14.40 |
| artificial analysis math index | 52.30 | 50.00 |
| gpqa | 62.0% | 40.8% |
| hle | 5.1% | 5.1% |
| ifbench | 31.2% | 25.5% |
| lcr | 0.0% | 0.0% |
| livecodebench | 42.2% | 49.3% |
| math 500 | — | 94.7% |
| mmlu pro | 72.5% | 55.6% |
| scicode | 18.6% | 10.1% |
| tau2 | 16.4% | 11.7% |
| terminalbench hard | 1.5% | — |
Benchmark data from Artificial Analysis.