← All comparisons

Llama Nemotron Super 49B v1.5 (Reasoning) vs Qwen3 235B A22B (Non-reasoning)

NVIDIA vs Alibaba — side-by-side benchmark comparison

Llama Nemotron Super 49B v1.5 (Reasoning)Qwen3 235B A22B (Non-reasoning)
Intelligence Index18.717.0
Coding Index15.114.0
Math Index76.723.7
Output speed (tok/s)48.160.8
Blended price ($/1M)$0.17$0.79
Time to first token (s)0.30s1.30s
aime86.0%32.7%
aime 2576.7%23.7%
artificial analysis coding index15.1014.00
artificial analysis intelligence index18.7017.00
artificial analysis math index76.7023.70
gpqa74.8%61.3%
hle6.8%4.7%
ifbench37.0%36.6%
lcr34.0%0.0%
livecodebench73.7%34.3%
math 50098.3%90.2%
mmlu pro81.4%76.2%
scicode34.8%29.9%
tau228.1%27.2%
terminalbench hard5.3%6.1%

Benchmark data from Artificial Analysis.