← All comparisons

Claude 4.1 Opus (Reasoning) vs DeepSeek V3.2 (Non-reasoning)

Anthropic vs DeepSeek — side-by-side benchmark comparison

Claude 4.1 Opus (Reasoning)DeepSeek V3.2 (Non-reasoning)
Intelligence Index42.032.1
Coding Index36.534.6
Math Index80.359.0
Output speed (tok/s)44.50.0
Blended price ($/1M)$32.81$0.78
Time to first token (s)8.55s0.00s
aime
aime 2580.3%59.0%
artificial analysis coding index36.5034.60
artificial analysis intelligence index42.0032.10
artificial analysis math index80.3059.00
gpqa80.9%75.1%
hle11.9%10.5%
ifbench55.4%49.0%
lcr66.3%39.0%
livecodebench65.4%59.3%
math 500
mmlu pro88.0%83.7%
scicode40.9%38.7%
tau271.4%78.9%
terminalbench hard34.3%32.6%

Benchmark data from Artificial Analysis.