← All comparisons

GPT-5.4 mini (Non-Reasoning) vs Claude 4.5 Haiku (Reasoning)

OpenAI vs Anthropic — side-by-side benchmark comparison

GPT-5.4 mini (Non-Reasoning)Claude 4.5 Haiku (Reasoning)
Intelligence Index23.337.1
Coding Index25.332.6
Math Index83.7
Output speed (tok/s)166.4142.2
Blended price ($/1M)$1.69$2.19
Time to first token (s)0.48s10.48s
aime
aime 2583.7%
artificial analysis coding index25.3032.60
artificial analysis intelligence index23.3037.10
artificial analysis math index83.70
gpqa60.6%67.2%
hle5.7%9.7%
ifbench38.8%54.3%
lcr30.3%70.3%
livecodebench61.5%
math 500
mmlu pro76.0%
scicode39.6%43.3%
tau223.4%54.7%
terminalbench hard18.2%27.3%

Benchmark data from Artificial Analysis.