← All comparisons

Claude 4.5 Haiku (Reasoning) vs GPT-4.1 mini

Anthropic vs OpenAI — side-by-side benchmark comparison

Claude 4.5 Haiku (Reasoning)GPT-4.1 mini
Intelligence Index37.122.9
Coding Index32.618.5
Math Index83.746.3
Output speed (tok/s)142.295.1
Blended price ($/1M)$2.19$0.70
Time to first token (s)10.48s0.59s
aime43.0%
aime 2583.7%46.3%
artificial analysis coding index32.6018.50
artificial analysis intelligence index37.1022.90
artificial analysis math index83.7046.30
gpqa67.2%66.4%
hle9.7%4.6%
ifbench54.3%38.3%
lcr70.3%42.3%
livecodebench61.5%48.3%
math 50092.5%
mmlu pro76.0%78.1%
scicode43.3%40.4%
tau254.7%52.9%
terminalbench hard27.3%7.6%

Benchmark data from Artificial Analysis.