← All comparisons
Grok 2 (Dec '24) vs Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)
xAI vs NVIDIA — side-by-side benchmark comparison
| Grok 2 (Dec '24) | Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 13.9 | 14.3 |
| Coding Index | — | 7.6 |
| Math Index | — | 7.7 |
| Output speed (tok/s) | 0.0 | 0.0 |
| Blended price ($/1M) | $0.00 | $0.00 |
| Time to first token (s) | 0.00s | 0.00s |
| aime | 13.3% | 19.3% |
| aime 25 | — | 7.7% |
| artificial analysis coding index | — | 7.60 |
| artificial analysis intelligence index | 13.90 | 14.30 |
| artificial analysis math index | — | 7.70 |
| gpqa | 51.0% | 51.7% |
| hle | 3.8% | 3.5% |
| ifbench | — | 39.5% |
| lcr | — | 11.3% |
| livecodebench | 26.7% | 28.0% |
| math 500 | 77.8% | 77.5% |
| mmlu pro | 70.9% | 69.8% |
| scicode | 28.5% | 22.9% |
| tau2 | — | — |
| terminalbench hard | — | 0.0% |
Benchmark data from Artificial Analysis.