← All comparisons

Claude 4.5 Haiku (Reasoning) vs MiniMax-M2.7

Anthropic vs MiniMax — side-by-side benchmark comparison

Claude 4.5 Haiku (Reasoning)MiniMax-M2.7
Intelligence Index37.149.6
Coding Index32.641.9
Math Index83.7
Output speed (tok/s)142.274.9
Blended price ($/1M)$2.19$0.53
Time to first token (s)10.48s2.46s
aime
aime 2583.7%
artificial analysis coding index32.6041.90
artificial analysis intelligence index37.1049.60
artificial analysis math index83.70
gpqa67.2%87.4%
hle9.7%28.1%
ifbench54.3%75.7%
lcr70.3%68.7%
livecodebench61.5%
math 500
mmlu pro76.0%
scicode43.3%47.0%
tau254.7%84.8%
terminalbench hard27.3%39.4%

Benchmark data from Artificial Analysis.