Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
Kimi K2 wins overall with a score of 23 vs 22 (1 point difference).Kimi K2 wins 4 out of 4 categories.
GPT-OSS 20B
28.8
Kimi K2
29.8
GPT-OSS 20B
23
Kimi K2
24
GPT-OSS 20B
30
Kimi K2
31
GPT-OSS 20B
28
Kimi K2
29
Kimi K2 scores higher overall with 23 vs 22, a difference of 1 points across all benchmarks.
Kimi K2 leads in knowledge tasks with an average score of 29.8 vs 28.8.
Kimi K2 leads in coding with an average score of 24 vs 23.
Kimi K2 leads in math with an average score of 31 vs 30.
Kimi K2 leads in reasoning with an average score of 29 vs 28.