← All comparisons

Gemini 3.1 Pro Preview vs Claude 4.5 Haiku (Non-reasoning)

Google vs Anthropic — side-by-side benchmark comparison

Gemini 3.1 Pro PreviewClaude 4.5 Haiku (Non-reasoning)
Intelligence Index57.231.0
Coding Index55.529.6
Math Index39.0
Output speed (tok/s)136.8105.8
Blended price ($/1M)$4.50$2.19
Time to first token (s)25.68s0.65s
aime
aime 2539.0%
artificial analysis coding index55.5029.60
artificial analysis intelligence index57.2031.00
artificial analysis math index39.00
gpqa94.1%64.6%
hle44.7%4.3%
ifbench77.1%42.0%
lcr72.7%43.7%
livecodebench51.1%
math 500
mmlu pro80.0%
scicode58.9%34.4%
tau295.6%32.5%
terminalbench hard53.8%27.3%

Benchmark data from Artificial Analysis.