← All comparisons

Claude 4.5 Haiku (Reasoning) vs GPT-5 (minimal)

Anthropic vs OpenAI — side-by-side benchmark comparison

Claude 4.5 Haiku (Reasoning)GPT-5 (minimal)
Intelligence Index37.123.9
Coding Index32.625.0
Math Index83.731.7
Output speed (tok/s)142.276.8
Blended price ($/1M)$2.19$3.44
Time to first token (s)10.48s0.84s
aime36.7%
aime 2583.7%31.7%
artificial analysis coding index32.6025.00
artificial analysis intelligence index37.1023.90
artificial analysis math index83.7031.70
gpqa67.2%67.3%
hle9.7%5.4%
ifbench54.3%45.6%
lcr70.3%25.0%
livecodebench61.5%55.8%
math 50086.1%
mmlu pro76.0%80.6%
scicode43.3%38.8%
tau254.7%67.0%
terminalbench hard27.3%18.2%

Benchmark data from Artificial Analysis.