← All comparisons

Claude 4.1 Opus (Reasoning) vs DeepSeek R1 0528 Qwen3 8B

Anthropic vs DeepSeek — side-by-side benchmark comparison

Claude 4.1 Opus (Reasoning)DeepSeek R1 0528 Qwen3 8B
Intelligence Index42.016.4
Coding Index36.57.8
Math Index80.363.7
Output speed (tok/s)44.50.0
Blended price ($/1M)$32.81$0.00
Time to first token (s)8.55s0.00s
aime65.0%
aime 2580.3%63.7%
artificial analysis coding index36.507.80
artificial analysis intelligence index42.0016.40
artificial analysis math index80.3063.70
gpqa80.9%61.2%
hle11.9%5.6%
ifbench55.4%19.9%
lcr66.3%13.0%
livecodebench65.4%51.3%
math 50093.2%
mmlu pro88.0%73.9%
scicode40.9%20.4%
tau271.4%0.0%
terminalbench hard34.3%1.5%

Benchmark data from Artificial Analysis.