GLM-4.5-Air vs Ministral 3 3B

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

GLM-4.5-Air is clearly ahead on the aggregate, 35 to 27. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

GLM-4.5-Air's sharpest advantage is in reasoning, where it averages 42.6 against 30.1. The single biggest benchmark swing on the page is MRCRv2, 51 to 35. Ministral 3 3B does hit back in multilingual, so the answer changes if that is the part of the workload you care about most.

Quick Verdict

Pick GLM-4.5-Air if you want the stronger benchmark profile. Ministral 3 3B only becomes the better choice if multilingual is the priority.

Agentic

GLM-4.5-Air

GLM-4.5-Air

30.3

Ministral 3 3B

22.9

28
Terminal-Bench 2.0
19
37
BrowseComp
33
28
OSWorld-Verified
20

Coding

GLM-4.5-Air

GLM-4.5-Air

14.6

Ministral 3 3B

6.2

27
HumanEval
15
15
SWE-bench Verified
8
15
LiveCodeBench
7
14
SWE-bench Pro
5

Multimodal & Grounded

GLM-4.5-Air

GLM-4.5-Air

39.6

Ministral 3 3B

30.4

36
MMMU-Pro
25
44
OfficeQA Pro
37

Reasoning

GLM-4.5-Air

GLM-4.5-Air

42.6

Ministral 3 3B

30.1

33
SimpleQA
22
31
MuSR
20
63
BBH
57
47
LongBench v2
32
51
MRCRv2
35

Knowledge

GLM-4.5-Air

GLM-4.5-Air

31.1

Ministral 3 3B

24.5

35
MMLU
24
34
GPQA
23
32
SuperGPQA
21
30
OpenBookQA
19
51
MMLU-Pro
48
4
HLE
1
37
FrontierScience
28

Instruction Following

GLM-4.5-Air

GLM-4.5-Air

68

Ministral 3 3B

67

68
IFEval
67

Multilingual

Ministral 3 3B

GLM-4.5-Air

59.1

Ministral 3 3B

59.7

63
MGSM
61
57
MMLU-ProX
59

Mathematics

GLM-4.5-Air

GLM-4.5-Air

44.4

Ministral 3 3B

36

35
AIME 2023
23
37
AIME 2024
25
36
AIME 2025
24
31
HMMT Feb 2023
19
33
HMMT Feb 2024
21
32
HMMT Feb 2025
20
34
BRUMO 2025
22
57
MATH-500
53

Frequently Asked Questions

Which is better, GLM-4.5-Air or Ministral 3 3B?

GLM-4.5-Air is ahead overall, 35 to 27. The biggest single separator in this matchup is MRCRv2, where the scores are 51 and 35.

Which is better for knowledge tasks, GLM-4.5-Air or Ministral 3 3B?

GLM-4.5-Air has the edge for knowledge tasks in this comparison, averaging 31.1 versus 24.5. Inside this category, MMLU is the benchmark that creates the most daylight between them.

Which is better for coding, GLM-4.5-Air or Ministral 3 3B?

GLM-4.5-Air has the edge for coding in this comparison, averaging 14.6 versus 6.2. Inside this category, HumanEval is the benchmark that creates the most daylight between them.

Which is better for math, GLM-4.5-Air or Ministral 3 3B?

GLM-4.5-Air has the edge for math in this comparison, averaging 44.4 versus 36. Inside this category, AIME 2023 is the benchmark that creates the most daylight between them.

Which is better for reasoning, GLM-4.5-Air or Ministral 3 3B?

GLM-4.5-Air has the edge for reasoning in this comparison, averaging 42.6 versus 30.1. Inside this category, MRCRv2 is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, GLM-4.5-Air or Ministral 3 3B?

GLM-4.5-Air has the edge for agentic tasks in this comparison, averaging 30.3 versus 22.9. 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, GLM-4.5-Air or Ministral 3 3B?

GLM-4.5-Air has the edge for multimodal and grounded tasks in this comparison, averaging 39.6 versus 30.4. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.

Which is better for instruction following, GLM-4.5-Air or Ministral 3 3B?

GLM-4.5-Air has the edge for instruction following in this comparison, averaging 68 versus 67. Inside this category, IFEval is the benchmark that creates the most daylight between them.

Which is better for multilingual tasks, GLM-4.5-Air or Ministral 3 3B?

Ministral 3 3B has the edge for multilingual tasks in this comparison, averaging 59.7 versus 59.1. 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.