← All comparisons

Claude 4.5 Haiku (Reasoning) vs Kimi K2 Thinking

Anthropic vs Kimi — side-by-side benchmark comparison

Claude 4.5 Haiku (Reasoning)Kimi K2 Thinking
Intelligence Index37.140.9
Coding Index32.634.8
Math Index83.794.7
Output speed (tok/s)142.2130.3
Blended price ($/1M)$2.19$1.07
Time to first token (s)10.48s0.84s
aime
aime 2583.7%94.7%
artificial analysis coding index32.6034.80
artificial analysis intelligence index37.1040.90
artificial analysis math index83.7094.70
gpqa67.2%83.8%
hle9.7%22.3%
ifbench54.3%68.1%
lcr70.3%66.3%
livecodebench61.5%85.3%
math 500
mmlu pro76.0%84.8%
scicode43.3%42.4%
tau254.7%93.0%
terminalbench hard27.3%31.1%

Benchmark data from Artificial Analysis.