benchlm.ai

Grok 3 [Beta] vs Kimi K2.5

Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.

Quick Verdict

Kimi K2.5 wins overall with a score of 62 vs 33 (29 point difference).Kimi K2.5 wins 4 out of 4 categories.

Grok 3 [Beta]

CreatorxAI
TypeProprietary
Context128K
Score33

Kimi K2.5

CreatorMoonshot AI
TypeOpen Weight
Context128K
Score62

Knowledge

Kimi K2.5

Grok 3 [Beta]

39.8

Kimi K2.5

74.8

42
MMLU
77
41
GPQA
76
39
SuperGPQA
74
37
OpenBookQA
72

Coding

Kimi K2.5

Grok 3 [Beta]

34

Kimi K2.5

69

34
HumanEval
69

Mathematics

Kimi K2.5

Grok 3 [Beta]

41

Kimi K2.5

76

42
AIME 2023
77
44
AIME 2024
79
43
AIME 2025
78
38
HMMT Feb 2023
73
40
HMMT Feb 2024
75
39
HMMT Feb 2025
74
41
BRUMO 2025
76

Reasoning

Kimi K2.5

Grok 3 [Beta]

39

Kimi K2.5

73

40
SimpleQA
74
38
MuSR
72

Frequently Asked Questions

Which is better, Grok 3 [Beta] or Kimi K2.5?

Kimi K2.5 scores higher overall with 62 vs 33, a difference of 29 points across all benchmarks.

Which is better for knowledge tasks, Grok 3 [Beta] or Kimi K2.5?

Kimi K2.5 leads in knowledge tasks with an average score of 74.8 vs 39.8.

Which is better for coding, Grok 3 [Beta] or Kimi K2.5?

Kimi K2.5 leads in coding with an average score of 69 vs 34.

Which is better for math, Grok 3 [Beta] or Kimi K2.5?

Kimi K2.5 leads in math with an average score of 76 vs 41.

Which is better for reasoning, Grok 3 [Beta] or Kimi K2.5?

Kimi K2.5 leads in reasoning with an average score of 73 vs 39.

Related Comparisons

Grok 3 [Beta] vs GPT-5.4Grok 3 [Beta] vs Gemini 3.1 ProGrok 3 [Beta] vs Claude Opus 4.6Grok 3 [Beta] vs GPT-5.3 CodexGrok 3 [Beta] vs Grok 4.1Grok 3 [Beta] vs GPT-5.2