← All comparisons

Gemini 2.5 Flash (Reasoning) vs Claude 3.5 Sonnet (Oct '24)

Google vs Anthropic — side-by-side benchmark comparison

Gemini 2.5 Flash (Reasoning)Claude 3.5 Sonnet (Oct '24)
Intelligence Index27.015.9
Coding Index22.230.2
Math Index73.3
Output speed (tok/s)205.50.0
Blended price ($/1M)$0.85$6.56
Time to first token (s)10.67s0.00s
aime82.3%15.7%
aime 2573.3%
artificial analysis coding index22.2030.20
artificial analysis intelligence index27.0015.90
artificial analysis math index73.30
gpqa79.0%59.9%
hle11.1%3.9%
ifbench50.3%
lcr61.7%
livecodebench69.5%38.1%
math 50098.1%77.1%
mmlu pro83.2%77.2%
scicode39.4%36.6%
tau231.6%
terminalbench hard13.6%

Benchmark data from Artificial Analysis.