← All comparisons

Claude 4.5 Haiku (Non-reasoning) vs DeepSeek V3.1 (Non-reasoning)

Anthropic vs DeepSeek — side-by-side benchmark comparison

Claude 4.5 Haiku (Non-reasoning)DeepSeek V3.1 (Non-reasoning)
Intelligence Index31.028.1
Coding Index29.628.4
Math Index39.049.7
Output speed (tok/s)105.80.0
Blended price ($/1M)$2.19$0.83
Time to first token (s)0.65s0.00s
aime
aime 2539.0%49.7%
artificial analysis coding index29.6028.40
artificial analysis intelligence index31.0028.10
artificial analysis math index39.0049.70
gpqa64.6%73.5%
hle4.3%6.3%
ifbench42.0%37.8%
lcr43.7%45.0%
livecodebench51.1%57.7%
math 500
mmlu pro80.0%83.3%
scicode34.4%36.7%
tau232.5%34.8%
terminalbench hard27.3%24.2%

Benchmark data from Artificial Analysis.