← All comparisons
DeepSeek V3.1 Terminus (Non-reasoning) vs Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)
DeepSeek vs NVIDIA — side-by-side benchmark comparison
| DeepSeek V3.1 Terminus (Non-reasoning) | Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 28.5 | 14.3 |
| Coding Index | 31.9 | 7.6 |
| Math Index | 53.7 | 7.7 |
| Output speed (tok/s) | 0.0 | 0.0 |
| Blended price ($/1M) | $0.45 | $0.00 |
| Time to first token (s) | 0.00s | 0.00s |
| aime | — | 19.3% |
| aime 25 | 53.7% | 7.7% |
| artificial analysis coding index | 31.90 | 7.60 |
| artificial analysis intelligence index | 28.50 | 14.30 |
| artificial analysis math index | 53.70 | 7.70 |
| gpqa | 75.1% | 51.7% |
| hle | 8.4% | 3.5% |
| ifbench | 41.2% | 39.5% |
| lcr | 43.3% | 11.3% |
| livecodebench | 52.9% | 28.0% |
| math 500 | — | 77.5% |
| mmlu pro | 83.6% | 69.8% |
| scicode | 32.1% | 22.9% |
| tau2 | 37.1% | — |
| terminalbench hard | 31.8% | 0.0% |
Benchmark data from Artificial Analysis.