← All comparisons
Llama Nemotron Super 49B v1.5 (Reasoning) vs Qwen3.5 35B A3B (Non-reasoning)
NVIDIA vs Alibaba — side-by-side benchmark comparison
| Llama Nemotron Super 49B v1.5 (Reasoning) | Qwen3.5 35B A3B (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 18.7 | 30.7 |
| Coding Index | 15.1 | 16.8 |
| Math Index | 76.7 | — |
| Output speed (tok/s) | 48.1 | 147.9 |
| Blended price ($/1M) | $0.17 | $0.69 |
| Time to first token (s) | 0.30s | 1.18s |
| aime | 86.0% | — |
| aime 25 | 76.7% | — |
| artificial analysis coding index | 15.10 | 16.80 |
| artificial analysis intelligence index | 18.70 | 30.70 |
| artificial analysis math index | 76.70 | — |
| gpqa | 74.8% | 81.9% |
| hle | 6.8% | 12.8% |
| ifbench | 37.0% | 44.5% |
| lcr | 34.0% | 55.3% |
| livecodebench | 73.7% | — |
| math 500 | 98.3% | — |
| mmlu pro | 81.4% | — |
| scicode | 34.8% | 29.3% |
| tau2 | 28.1% | 86.3% |
| terminalbench hard | 5.3% | 10.6% |
Benchmark data from Artificial Analysis.