← All comparisons
Llama Nemotron Super 49B v1.5 (Reasoning) vs Qwen3 235B A22B (Non-reasoning)
NVIDIA vs Alibaba — side-by-side benchmark comparison
| Llama Nemotron Super 49B v1.5 (Reasoning) | Qwen3 235B A22B (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 18.7 | 17.0 |
| Coding Index | 15.1 | 14.0 |
| Math Index | 76.7 | 23.7 |
| Output speed (tok/s) | 48.1 | 60.8 |
| Blended price ($/1M) | $0.17 | $0.79 |
| Time to first token (s) | 0.30s | 1.30s |
| aime | 86.0% | 32.7% |
| aime 25 | 76.7% | 23.7% |
| artificial analysis coding index | 15.10 | 14.00 |
| artificial analysis intelligence index | 18.70 | 17.00 |
| artificial analysis math index | 76.70 | 23.70 |
| gpqa | 74.8% | 61.3% |
| hle | 6.8% | 4.7% |
| ifbench | 37.0% | 36.6% |
| lcr | 34.0% | 0.0% |
| livecodebench | 73.7% | 34.3% |
| math 500 | 98.3% | 90.2% |
| mmlu pro | 81.4% | 76.2% |
| scicode | 34.8% | 29.9% |
| tau2 | 28.1% | 27.2% |
| terminalbench hard | 5.3% | 6.1% |
Benchmark data from Artificial Analysis.