← All comparisons

Claude 4.5 Haiku (Non-reasoning) vs MiMo-V2.5-Pro (Non-reasoning)

Anthropic vs Xiaomi — side-by-side benchmark comparison

Claude 4.5 Haiku (Non-reasoning)MiMo-V2.5-Pro (Non-reasoning)
Intelligence Index31.035.6
Coding Index29.636.8
Math Index39.0
Output speed (tok/s)105.853.5
Blended price ($/1M)$2.19$1.35
Time to first token (s)0.65s2.13s
aime
aime 2539.0%
artificial analysis coding index29.6036.80
artificial analysis intelligence index31.0035.60
artificial analysis math index39.00
gpqa64.6%76.2%
hle4.3%13.3%
ifbench42.0%42.7%
lcr43.7%35.0%
livecodebench51.1%
math 500
mmlu pro80.0%
scicode34.4%39.1%
tau232.5%72.5%
terminalbench hard27.3%35.6%

Benchmark data from Artificial Analysis.