← All comparisons

Claude 4.5 Haiku (Reasoning) vs Gemini 2.5 Flash (Reasoning)

Anthropic vs Google — side-by-side benchmark comparison

Claude 4.5 Haiku (Reasoning)Gemini 2.5 Flash (Reasoning)
Intelligence Index37.127.0
Coding Index32.622.2
Math Index83.773.3
Output speed (tok/s)142.2205.5
Blended price ($/1M)$2.19$0.85
Time to first token (s)10.48s10.67s
aime82.3%
aime 2583.7%73.3%
artificial analysis coding index32.6022.20
artificial analysis intelligence index37.1027.00
artificial analysis math index83.7073.30
gpqa67.2%79.0%
hle9.7%11.1%
ifbench54.3%50.3%
lcr70.3%61.7%
livecodebench61.5%69.5%
math 50098.1%
mmlu pro76.0%83.2%
scicode43.3%39.4%
tau254.7%31.6%
terminalbench hard27.3%13.6%

Benchmark data from Artificial Analysis.