← All comparisons

K2 Think V2 vs Claude 4.5 Sonnet (Reasoning)

MBZUAI Institute of Foundation Models vs Anthropic — side-by-side benchmark comparison

K2 Think V2Claude 4.5 Sonnet (Reasoning)
Intelligence Index24.143.0
Coding Index15.538.6
Math Index88.0
Output speed (tok/s)0.055.0
Blended price ($/1M)$0.00$6.56
Time to first token (s)0.00s7.02s
aime
aime 2588.0%
artificial analysis coding index15.5038.60
artificial analysis intelligence index24.1043.00
artificial analysis math index88.00
gpqa71.3%83.4%
hle9.5%17.3%
ifbench62.8%57.3%
lcr52.7%65.7%
livecodebench71.4%
math 500
mmlu pro87.5%
scicode33.0%44.7%
tau225.4%78.1%
terminalbench hard6.8%35.6%

Benchmark data from Artificial Analysis.