← All comparisons

Claude 4.5 Sonnet (Reasoning) vs DeepSeek R1 Distill Qwen 14B

Anthropic vs DeepSeek — side-by-side benchmark comparison

Claude 4.5 Sonnet (Reasoning)DeepSeek R1 Distill Qwen 14B
Intelligence Index43.015.8
Coding Index38.6
Math Index88.055.7
Output speed (tok/s)55.00.0
Blended price ($/1M)$6.56$0.00
Time to first token (s)7.02s0.00s
aime66.7%
aime 2588.0%55.7%
artificial analysis coding index38.60
artificial analysis intelligence index43.0015.80
artificial analysis math index88.0055.70
gpqa83.4%48.4%
hle17.3%4.4%
ifbench57.3%22.1%
lcr65.7%7.0%
livecodebench71.4%37.6%
math 50094.9%
mmlu pro87.5%74.0%
scicode44.7%23.9%
tau278.1%
terminalbench hard35.6%

Benchmark data from Artificial Analysis.