← All comparisons

Llama 3.2 Instruct 3B vs Claude 4.1 Opus (Non-reasoning)

Meta vs Anthropic — side-by-side benchmark comparison

Llama 3.2 Instruct 3BClaude 4.1 Opus (Non-reasoning)
Intelligence Index9.736.0
Coding Index
Math Index3.3
Output speed (tok/s)52.744.7
Blended price ($/1M)$0.15$32.81
Time to first token (s)0.65s1.63s
aime6.7%
aime 253.3%
artificial analysis coding index
artificial analysis intelligence index9.7036.00
artificial analysis math index3.30
gpqa25.5%
hle5.2%
ifbench26.2%
lcr2.0%
livecodebench8.3%
math 50048.9%
mmlu pro34.7%
scicode5.2%
tau221.1%
terminalbench hard

Benchmark data from Artificial Analysis.