← All comparisons

Gemini 3 Pro Preview (high) vs Claude 4.5 Sonnet (Reasoning)

Google vs Anthropic — side-by-side benchmark comparison

Gemini 3 Pro Preview (high)Claude 4.5 Sonnet (Reasoning)
Intelligence Index48.443.0
Coding Index46.538.6
Math Index95.788.0
Output speed (tok/s)0.055.0
Blended price ($/1M)$4.50$6.56
Time to first token (s)0.00s7.02s
aime
aime 2595.7%88.0%
artificial analysis coding index46.5038.60
artificial analysis intelligence index48.4043.00
artificial analysis math index95.7088.00
gpqa90.8%83.4%
hle37.2%17.3%
ifbench70.4%57.3%
lcr70.7%65.7%
livecodebench91.7%71.4%
math 500
mmlu pro89.8%87.5%
scicode56.1%44.7%
tau287.1%78.1%
terminalbench hard41.7%35.6%

Benchmark data from Artificial Analysis.