← All comparisons

Claude 4.5 Haiku (Non-reasoning) vs Gemini 2.5 Flash-Lite (Reasoning)

Anthropic vs Google — side-by-side benchmark comparison

Claude 4.5 Haiku (Non-reasoning)Gemini 2.5 Flash-Lite (Reasoning)
Intelligence Index31.017.6
Coding Index29.69.5
Math Index39.053.3
Output speed (tok/s)105.8275.9
Blended price ($/1M)$2.19$0.17
Time to first token (s)0.65s17.31s
aime70.3%
aime 2539.0%53.3%
artificial analysis coding index29.609.50
artificial analysis intelligence index31.0017.60
artificial analysis math index39.0053.30
gpqa64.6%62.5%
hle4.3%6.4%
ifbench42.0%49.9%
lcr43.7%51.3%
livecodebench51.1%59.3%
math 50096.9%
mmlu pro80.0%75.9%
scicode34.4%19.3%
tau232.5%18.4%
terminalbench hard27.3%4.5%

Benchmark data from Artificial Analysis.