Mistral Large 2 vs Ministral 3 3B

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

Mistral Large 2 is clearly ahead on the aggregate, 59 to 27. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Mistral Large 2's sharpest advantage is in mathematics, where it averages 73.8 against 36. The single biggest benchmark swing on the page is HumanEval, 60 to 15.

Quick Verdict

Pick Mistral Large 2 if you want the stronger benchmark profile. Ministral 3 3B only becomes the better choice if its workflow or ecosystem matters more than the raw scoreboard.

Agentic

Mistral Large 2

Mistral Large 2

52.2

Ministral 3 3B

22.9

51
Terminal-Bench 2.0
19
57
BrowseComp
33
50
OSWorld-Verified
20

Coding

Mistral Large 2

Mistral Large 2

41.8

Ministral 3 3B

6.2

60
HumanEval
15
49
SWE-bench Verified
8
38
LiveCodeBench
7
44
SWE-bench Pro
5

Multimodal & Grounded

Mistral Large 2

Mistral Large 2

61

Ministral 3 3B

30.4

56
MMMU-Pro
25
67
OfficeQA Pro
37

Reasoning

Mistral Large 2

Mistral Large 2

67.5

Ministral 3 3B

30.1

66
SimpleQA
22
64
MuSR
20
82
BBH
57
66
LongBench v2
32
68
MRCRv2
35

Knowledge

Mistral Large 2

Mistral Large 2

54.6

Ministral 3 3B

24.5

68
MMLU
24
68
GPQA
23
66
SuperGPQA
21
64
OpenBookQA
19
74
MMLU-Pro
48
12
HLE
1
65
FrontierScience
28

Instruction Following

Mistral Large 2

Mistral Large 2

83

Ministral 3 3B

67

83
IFEval
67

Multilingual

Mistral Large 2

Mistral Large 2

79.1

Ministral 3 3B

59.7

81
MGSM
61
78
MMLU-ProX
59

Mathematics

Mistral Large 2

Mistral Large 2

73.8

Ministral 3 3B

36

68
AIME 2023
23
70
AIME 2024
25
69
AIME 2025
24
64
HMMT Feb 2023
19
66
HMMT Feb 2024
21
65
HMMT Feb 2025
20
67
BRUMO 2025
22
82
MATH-500
53

Frequently Asked Questions

Which is better, Mistral Large 2 or Ministral 3 3B?

Mistral Large 2 is ahead overall, 59 to 27. The biggest single separator in this matchup is HumanEval, where the scores are 60 and 15.

Which is better for knowledge tasks, Mistral Large 2 or Ministral 3 3B?

Mistral Large 2 has the edge for knowledge tasks in this comparison, averaging 54.6 versus 24.5. Inside this category, GPQA is the benchmark that creates the most daylight between them.

Which is better for coding, Mistral Large 2 or Ministral 3 3B?

Mistral Large 2 has the edge for coding in this comparison, averaging 41.8 versus 6.2. Inside this category, HumanEval is the benchmark that creates the most daylight between them.

Which is better for math, Mistral Large 2 or Ministral 3 3B?

Mistral Large 2 has the edge for math in this comparison, averaging 73.8 versus 36. Inside this category, AIME 2023 is the benchmark that creates the most daylight between them.

Which is better for reasoning, Mistral Large 2 or Ministral 3 3B?

Mistral Large 2 has the edge for reasoning in this comparison, averaging 67.5 versus 30.1. Inside this category, SimpleQA is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Mistral Large 2 or Ministral 3 3B?

Mistral Large 2 has the edge for agentic tasks in this comparison, averaging 52.2 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, Mistral Large 2 or Ministral 3 3B?

Mistral Large 2 has the edge for multimodal and grounded tasks in this comparison, averaging 61 versus 30.4. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.

Which is better for instruction following, Mistral Large 2 or Ministral 3 3B?

Mistral Large 2 has the edge for instruction following in this comparison, averaging 83 versus 67. Inside this category, IFEval is the benchmark that creates the most daylight between them.

Which is better for multilingual tasks, Mistral Large 2 or Ministral 3 3B?

Mistral Large 2 has the edge for multilingual tasks in this comparison, averaging 79.1 versus 59.7. 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.