← All comparisons

Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) vs Qwen3 VL 8B Instruct

NVIDIA vs Alibaba — side-by-side benchmark comparison

Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)Qwen3 VL 8B Instruct
Intelligence Index15.014.3
Coding Index13.17.3
Math Index63.727.3
Output speed (tok/s)52.3143.8
Blended price ($/1M)$0.90$0.31
Time to first token (s)0.76s0.93s
aime74.7%
aime 2563.7%27.3%
artificial analysis coding index13.107.30
artificial analysis intelligence index15.0014.30
artificial analysis math index63.7027.30
gpqa72.8%42.7%
hle8.1%2.9%
ifbench38.2%32.3%
lcr7.3%15.3%
livecodebench64.1%33.2%
math 50095.2%
mmlu pro82.5%68.6%
scicode34.7%17.4%
tau211.4%29.2%
terminalbench hard2.3%2.3%

Benchmark data from Artificial Analysis.