← All comparisons

Gemma 4 E4B (Reasoning) vs Claude 4.5 Haiku (Non-reasoning)

Google vs Anthropic — side-by-side benchmark comparison

Gemma 4 E4B (Reasoning)Claude 4.5 Haiku (Non-reasoning)
Intelligence Index18.831.0
Coding Index13.729.6
Math Index39.0
Output speed (tok/s)0.0105.8
Blended price ($/1M)$0.00$2.19
Time to first token (s)0.00s0.65s
aime
aime 2539.0%
artificial analysis coding index13.7029.60
artificial analysis intelligence index18.8031.00
artificial analysis math index39.00
gpqa57.6%64.6%
hle3.7%4.3%
ifbench44.2%42.0%
lcr30.7%43.7%
livecodebench51.1%
math 500
mmlu pro80.0%
scicode24.4%34.4%
tau220.8%32.5%
terminalbench hard8.3%27.3%

Benchmark data from Artificial Analysis.