← All comparisons
Llama Nemotron Super 49B v1.5 (Non-reasoning) vs Claude 4.1 Opus (Reasoning)
NVIDIA vs Anthropic — side-by-side benchmark comparison
| Llama Nemotron Super 49B v1.5 (Non-reasoning) | Claude 4.1 Opus (Reasoning) | |
|---|---|---|
| Intelligence Index | 14.6 | 42.0 |
| Coding Index | 10.5 | 36.5 |
| Math Index | 8.0 | 80.3 |
| Output speed (tok/s) | 48.9 | 44.5 |
| Blended price ($/1M) | $0.17 | $32.81 |
| Time to first token (s) | 0.28s | 8.55s |
| aime | 13.7% | — |
| aime 25 | 8.0% | 80.3% |
| artificial analysis coding index | 10.50 | 36.50 |
| artificial analysis intelligence index | 14.60 | 42.00 |
| artificial analysis math index | 8.00 | 80.30 |
| gpqa | 48.1% | 80.9% |
| hle | 4.3% | 11.9% |
| ifbench | 32.9% | 55.4% |
| lcr | 22.0% | 66.3% |
| livecodebench | 29.0% | 65.4% |
| math 500 | 77.0% | — |
| mmlu pro | 69.2% | 88.0% |
| scicode | 23.8% | 40.9% |
| tau2 | 25.1% | 71.4% |
| terminalbench hard | 3.8% | 34.3% |
Benchmark data from Artificial Analysis.