← All comparisons

Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) vs Claude 4.1 Opus (Reasoning)

NVIDIA vs Anthropic — side-by-side benchmark comparison

Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)Claude 4.1 Opus (Reasoning)
Intelligence Index15.042.0
Coding Index13.136.5
Math Index63.780.3
Output speed (tok/s)52.344.5
Blended price ($/1M)$0.90$32.81
Time to first token (s)0.76s8.55s
aime74.7%
aime 2563.7%80.3%
artificial analysis coding index13.1036.50
artificial analysis intelligence index15.0042.00
artificial analysis math index63.7080.30
gpqa72.8%80.9%
hle8.1%11.9%
ifbench38.2%55.4%
lcr7.3%66.3%
livecodebench64.1%65.4%
math 50095.2%
mmlu pro82.5%88.0%
scicode34.7%40.9%
tau211.4%71.4%
terminalbench hard2.3%34.3%

Benchmark data from Artificial Analysis.