Kimi K2.5 vs Ministral 3 8B

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

Kimi K2.5 is clearly ahead on the aggregate, 60 to 32. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Kimi K2.5's sharpest advantage is in reasoning, where it averages 71.7 against 36.1. The single biggest benchmark swing on the page is MMLU, 77 to 28.

Kimi K2.5 is also the more expensive model on tokens at $0.50 input / $2.80 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Ministral 3 8B. That is roughly Infinityx on output cost alone.

Quick Verdict

Pick Kimi K2.5 if you want the stronger benchmark profile. Ministral 3 8B only becomes the better choice if you want the cheaper token bill.

Agentic

Kimi K2.5

Kimi K2.5

52.3

Ministral 3 8B

28.9

51
Terminal-Bench 2.0
26
59
BrowseComp
36
49
OSWorld-Verified
27

Coding

Kimi K2.5

Kimi K2.5

38.9

Ministral 3 8B

14.2

69
HumanEval
23
42
SWE-bench Verified
16
37
LiveCodeBench
15
40
SWE-bench Pro
13

Multimodal & Grounded

Kimi K2.5

Kimi K2.5

64.6

Ministral 3 8B

32.4

61
MMMU-Pro
27
69
OfficeQA Pro
39

Reasoning

Kimi K2.5

Kimi K2.5

71.7

Ministral 3 8B

36.1

74
SimpleQA
28
72
MuSR
26
81
BBH
63
67
LongBench v2
38
70
MRCRv2
41

Knowledge

Kimi K2.5

Kimi K2.5

57.2

Ministral 3 8B

28

77
MMLU
28
76
GPQA
27
74
SuperGPQA
25
72
OpenBookQA
23
74
MMLU-Pro
52
11
HLE
3
67
FrontierScience
32

Instruction Following

Kimi K2.5

Kimi K2.5

85

Ministral 3 8B

69

85
IFEval
69

Multilingual

Kimi K2.5

Kimi K2.5

79.8

Ministral 3 8B

61.7

83
MGSM
63
78
MMLU-ProX
61

Mathematics

Kimi K2.5

Kimi K2.5

78.7

Ministral 3 8B

43.3

77
AIME 2023
30
79
AIME 2024
32
78
AIME 2025
33
73
HMMT Feb 2023
26
75
HMMT Feb 2024
28
74
HMMT Feb 2025
27
76
BRUMO 2025
29
82
MATH-500
60

Frequently Asked Questions

Which is better, Kimi K2.5 or Ministral 3 8B?

Kimi K2.5 is ahead overall, 60 to 32. The biggest single separator in this matchup is MMLU, where the scores are 77 and 28.

Which is better for knowledge tasks, Kimi K2.5 or Ministral 3 8B?

Kimi K2.5 has the edge for knowledge tasks in this comparison, averaging 57.2 versus 28. Inside this category, MMLU is the benchmark that creates the most daylight between them.

Which is better for coding, Kimi K2.5 or Ministral 3 8B?

Kimi K2.5 has the edge for coding in this comparison, averaging 38.9 versus 14.2. Inside this category, HumanEval is the benchmark that creates the most daylight between them.

Which is better for math, Kimi K2.5 or Ministral 3 8B?

Kimi K2.5 has the edge for math in this comparison, averaging 78.7 versus 43.3. Inside this category, AIME 2023 is the benchmark that creates the most daylight between them.

Which is better for reasoning, Kimi K2.5 or Ministral 3 8B?

Kimi K2.5 has the edge for reasoning in this comparison, averaging 71.7 versus 36.1. Inside this category, SimpleQA is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Kimi K2.5 or Ministral 3 8B?

Kimi K2.5 has the edge for agentic tasks in this comparison, averaging 52.3 versus 28.9. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.

Which is better for multimodal and grounded tasks, Kimi K2.5 or Ministral 3 8B?

Kimi K2.5 has the edge for multimodal and grounded tasks in this comparison, averaging 64.6 versus 32.4. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.

Which is better for instruction following, Kimi K2.5 or Ministral 3 8B?

Kimi K2.5 has the edge for instruction following in this comparison, averaging 85 versus 69. Inside this category, IFEval is the benchmark that creates the most daylight between them.

Which is better for multilingual tasks, Kimi K2.5 or Ministral 3 8B?

Kimi K2.5 has the edge for multilingual tasks in this comparison, averaging 79.8 versus 61.7. Inside this category, MGSM is the benchmark that creates the most daylight between them.

Last updated: March 12, 2026

Weekly LLM Benchmark Digest

Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.

Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.