← All comparisons

Claude 4.1 Opus (Reasoning) vs DeepSeek R1 (Jan '25)

Anthropic vs DeepSeek — side-by-side benchmark comparison

Claude 4.1 Opus (Reasoning)DeepSeek R1 (Jan '25)
Intelligence Index42.018.8
Coding Index36.515.9
Math Index80.368.0
Output speed (tok/s)44.50.0
Blended price ($/1M)$32.81$2.43
Time to first token (s)8.55s0.00s
aime68.3%
aime 2580.3%68.0%
artificial analysis coding index36.5015.90
artificial analysis intelligence index42.0018.80
artificial analysis math index80.3068.00
gpqa80.9%70.8%
hle11.9%9.3%
ifbench55.4%39.0%
lcr66.3%52.3%
livecodebench65.4%61.7%
math 50096.6%
mmlu pro88.0%84.4%
scicode40.9%35.7%
tau271.4%11.4%
terminalbench hard34.3%6.1%

Benchmark data from Artificial Analysis.