← All comparisons

DeepSeek V4 Pro (Reasoning, Max Effort) vs Claude 4.1 Opus (Reasoning)

DeepSeek vs Anthropic — side-by-side benchmark comparison

DeepSeek V4 Pro (Reasoning, Max Effort)Claude 4.1 Opus (Reasoning)
Intelligence Index51.542.0
Coding Index47.536.5
Math Index80.3
Output speed (tok/s)51.844.5
Blended price ($/1M)$0.54$32.81
Time to first token (s)1.21s8.55s
aime
aime 2580.3%
artificial analysis coding index47.5036.50
artificial analysis intelligence index51.5042.00
artificial analysis math index80.30
gpqa88.8%80.9%
hle35.9%11.9%
ifbench76.5%55.4%
lcr66.3%66.3%
livecodebench65.4%
math 500
mmlu pro88.0%
scicode50.0%40.9%
tau296.2%71.4%
terminalbench hard46.2%34.3%

Benchmark data from Artificial Analysis.