← All comparisons

Claude 4.5 Haiku (Reasoning) vs Kimi K2.6 (Non-reasoning)

Anthropic vs Kimi — side-by-side benchmark comparison

Claude 4.5 Haiku (Reasoning)Kimi K2.6 (Non-reasoning)
Intelligence Index37.142.9
Coding Index32.638.4
Math Index83.7
Output speed (tok/s)142.239.0
Blended price ($/1M)$2.19$1.71
Time to first token (s)10.48s1.28s
aime
aime 2583.7%
artificial analysis coding index32.6038.40
artificial analysis intelligence index37.1042.90
artificial analysis math index83.70
gpqa67.2%78.8%
hle9.7%18.2%
ifbench54.3%44.3%
lcr70.3%57.7%
livecodebench61.5%
math 500
mmlu pro76.0%
scicode43.3%39.5%
tau254.7%93.9%
terminalbench hard27.3%37.9%

Benchmark data from Artificial Analysis.