← All comparisons

Claude 4.5 Haiku (Reasoning) vs Kimi Linear 48B A3B Instruct

Anthropic vs Kimi — side-by-side benchmark comparison

Claude 4.5 Haiku (Reasoning)Kimi Linear 48B A3B Instruct
Intelligence Index37.114.4
Coding Index32.614.2
Math Index83.736.3
Output speed (tok/s)142.20.0
Blended price ($/1M)$2.19$0.00
Time to first token (s)10.48s0.00s
aime
aime 2583.7%36.3%
artificial analysis coding index32.6014.20
artificial analysis intelligence index37.1014.40
artificial analysis math index83.7036.30
gpqa67.2%41.2%
hle9.7%2.7%
ifbench54.3%28.1%
lcr70.3%25.7%
livecodebench61.5%37.8%
math 500
mmlu pro76.0%58.5%
scicode43.3%19.9%
tau254.7%0.0%
terminalbench hard27.3%11.4%

Benchmark data from Artificial Analysis.