← All comparisons

Llama Nemotron Super 49B v1.5 (Reasoning) vs DeepSeek R1 Distill Qwen 32B

NVIDIA vs DeepSeek — side-by-side benchmark comparison

Llama Nemotron Super 49B v1.5 (Reasoning)DeepSeek R1 Distill Qwen 32B
Intelligence Index18.717.2
Coding Index15.1
Math Index76.763.0
Output speed (tok/s)48.10.0
Blended price ($/1M)$0.17$0.00
Time to first token (s)0.30s0.00s
aime86.0%68.7%
aime 2576.7%63.0%
artificial analysis coding index15.10
artificial analysis intelligence index18.7017.20
artificial analysis math index76.7063.00
gpqa74.8%61.5%
hle6.8%5.5%
ifbench37.0%22.9%
lcr34.0%9.7%
livecodebench73.7%27.0%
math 50098.3%94.1%
mmlu pro81.4%73.9%
scicode34.8%37.6%
tau228.1%
terminalbench hard5.3%

Benchmark data from Artificial Analysis.