Ministral 3 14B vs GPT-OSS 20B

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

Ministral 3 14B is clearly ahead on the aggregate, 55 to 35. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Ministral 3 14B's sharpest advantage is in multimodal & grounded, where it averages 70.5 against 36. The single biggest benchmark swing on the page is AIME 2025, 72 to 32.

Quick Verdict

Pick Ministral 3 14B if you want the stronger benchmark profile. GPT-OSS 20B only becomes the better choice if its workflow or ecosystem matters more than the raw scoreboard.

Agentic

Ministral 3 14B

Ministral 3 14B

48.4

GPT-OSS 20B

35.4

48
Terminal-Bench 2.0
35
55
BrowseComp
42
44
OSWorld-Verified
31

Coding

Ministral 3 14B

Ministral 3 14B

33

GPT-OSS 20B

14.5

58
HumanEval
23
37
SWE-bench Verified
14
31
LiveCodeBench
11
34
SWE-bench Pro
18

Multimodal & Grounded

Ministral 3 14B

Ministral 3 14B

70.5

GPT-OSS 20B

36

70
MMMU-Pro
31
71
OfficeQA Pro
42

Reasoning

Ministral 3 14B

Ministral 3 14B

63.6

GPT-OSS 20B

40.4

66
SimpleQA
29
64
MuSR
27
74
BBH
62
60
LongBench v2
48
60
MRCRv2
48

Knowledge

Ministral 3 14B

Ministral 3 14B

50.1

GPT-OSS 20B

29

69
MMLU
31
68
GPQA
30
66
SuperGPQA
28
64
OpenBookQA
26
67
MMLU-Pro
53
5
HLE
1
60
FrontierScience
34

Instruction Following

Ministral 3 14B

Ministral 3 14B

80

GPT-OSS 20B

67

80
IFEval
67

Multilingual

Ministral 3 14B

Ministral 3 14B

76.8

GPT-OSS 20B

59.7

80
MGSM
61
75
MMLU-ProX
59

Mathematics

Ministral 3 14B

Ministral 3 14B

69.7

GPT-OSS 20B

43.1

68
AIME 2023
31
70
AIME 2024
33
72
AIME 2025
32
64
HMMT Feb 2023
27
66
HMMT Feb 2024
29
65
HMMT Feb 2025
28
67
BRUMO 2025
30
72
MATH-500
59

Frequently Asked Questions

Which is better, Ministral 3 14B or GPT-OSS 20B?

Ministral 3 14B is ahead overall, 55 to 35. The biggest single separator in this matchup is AIME 2025, where the scores are 72 and 32.

Which is better for knowledge tasks, Ministral 3 14B or GPT-OSS 20B?

Ministral 3 14B has the edge for knowledge tasks in this comparison, averaging 50.1 versus 29. Inside this category, MMLU is the benchmark that creates the most daylight between them.

Which is better for coding, Ministral 3 14B or GPT-OSS 20B?

Ministral 3 14B has the edge for coding in this comparison, averaging 33 versus 14.5. Inside this category, HumanEval is the benchmark that creates the most daylight between them.

Which is better for math, Ministral 3 14B or GPT-OSS 20B?

Ministral 3 14B has the edge for math in this comparison, averaging 69.7 versus 43.1. Inside this category, AIME 2025 is the benchmark that creates the most daylight between them.

Which is better for reasoning, Ministral 3 14B or GPT-OSS 20B?

Ministral 3 14B has the edge for reasoning in this comparison, averaging 63.6 versus 40.4. Inside this category, SimpleQA is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Ministral 3 14B or GPT-OSS 20B?

Ministral 3 14B has the edge for agentic tasks in this comparison, averaging 48.4 versus 35.4. 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, Ministral 3 14B or GPT-OSS 20B?

Ministral 3 14B has the edge for multimodal and grounded tasks in this comparison, averaging 70.5 versus 36. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.

Which is better for instruction following, Ministral 3 14B or GPT-OSS 20B?

Ministral 3 14B has the edge for instruction following in this comparison, averaging 80 versus 67. Inside this category, IFEval is the benchmark that creates the most daylight between them.

Which is better for multilingual tasks, Ministral 3 14B or GPT-OSS 20B?

Ministral 3 14B has the edge for multilingual tasks in this comparison, averaging 76.8 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.