← All comparisons

Gemma 4 31B (Reasoning) vs Claude 4.5 Haiku (Reasoning)

Google vs Anthropic — side-by-side benchmark comparison

Gemma 4 31B (Reasoning)Claude 4.5 Haiku (Reasoning)
Intelligence Index39.237.1
Coding Index38.732.6
Math Index83.7
Output speed (tok/s)35.3142.2
Blended price ($/1M)$0.00$2.19
Time to first token (s)1.00s10.48s
aime
aime 2583.7%
artificial analysis coding index38.7032.60
artificial analysis intelligence index39.2037.10
artificial analysis math index83.70
gpqa85.7%67.2%
hle22.7%9.7%
ifbench75.6%54.3%
lcr62.0%70.3%
livecodebench61.5%
math 500
mmlu pro76.0%
scicode43.4%43.3%
tau259.9%54.7%
terminalbench hard36.4%27.3%

Benchmark data from Artificial Analysis.