← All comparisons

Claude 4.1 Opus (Non-reasoning) vs DeepSeek V3 (Dec '24)

Anthropic vs DeepSeek — side-by-side benchmark comparison

Claude 4.1 Opus (Non-reasoning)DeepSeek V3 (Dec '24)
Intelligence Index36.016.5
Coding Index16.4
Math Index26.0
Output speed (tok/s)44.70.0
Blended price ($/1M)$32.81$0.52
Time to first token (s)1.63s0.00s
aime25.3%
aime 2526.0%
artificial analysis coding index16.40
artificial analysis intelligence index36.0016.50
artificial analysis math index26.00
gpqa55.7%
hle3.6%
ifbench34.8%
lcr29.0%
livecodebench35.9%
math 50088.7%
mmlu pro75.2%
scicode35.4%
tau222.8%
terminalbench hard6.8%

Benchmark data from Artificial Analysis.