← All comparisons

Claude 4.5 Haiku (Reasoning) vs GPT-5 mini (high)

Anthropic vs OpenAI — side-by-side benchmark comparison

Claude 4.5 Haiku (Reasoning)GPT-5 mini (high)
Intelligence Index37.141.2
Coding Index32.635.3
Math Index83.790.7
Output speed (tok/s)142.296.1
Blended price ($/1M)$2.19$0.69
Time to first token (s)10.48s77.65s
aime
aime 2583.7%90.7%
artificial analysis coding index32.6035.30
artificial analysis intelligence index37.1041.20
artificial analysis math index83.7090.70
gpqa67.2%82.8%
hle9.7%19.7%
ifbench54.3%75.4%
lcr70.3%68.0%
livecodebench61.5%83.8%
math 500
mmlu pro76.0%83.7%
scicode43.3%39.2%
tau254.7%68.4%
terminalbench hard27.3%33.3%

Benchmark data from Artificial Analysis.