← All comparisons
Llama Nemotron Super 49B v1.5 (Non-reasoning) vs Qwen3.5 27B (Reasoning)
NVIDIA vs Alibaba — side-by-side benchmark comparison
| Llama Nemotron Super 49B v1.5 (Non-reasoning) | Qwen3.5 27B (Reasoning) | |
|---|---|---|
| Intelligence Index | 14.6 | 42.1 |
| Coding Index | 10.5 | 34.9 |
| Math Index | 8.0 | — |
| Output speed (tok/s) | 48.9 | 84.3 |
| Blended price ($/1M) | $0.17 | $0.82 |
| Time to first token (s) | 0.28s | 1.37s |
| aime | 13.7% | — |
| aime 25 | 8.0% | — |
| artificial analysis coding index | 10.50 | 34.90 |
| artificial analysis intelligence index | 14.60 | 42.10 |
| artificial analysis math index | 8.00 | — |
| gpqa | 48.1% | 85.8% |
| hle | 4.3% | 22.2% |
| ifbench | 32.9% | 75.6% |
| lcr | 22.0% | 67.3% |
| livecodebench | 29.0% | — |
| math 500 | 77.0% | — |
| mmlu pro | 69.2% | — |
| scicode | 23.8% | 39.5% |
| tau2 | 25.1% | 93.9% |
| terminalbench hard | 3.8% | 32.6% |
Benchmark data from Artificial Analysis.