Ministral 3 3B Benchmark Scores & Performance

Benchmark analysis of Ministral 3 3B by Mistral across 32 sourced tests on BenchLM.

According to BenchLM.ai, Ministral 3 3B ranks #121 out of 123 models with an overall score of 27/100. While not a frontier model, it offers specific advantages depending on the use case.

Ministral 3 3B is a open weight model with a 128K token context window. It processes queries without explicit chain-of-thought reasoning, offering faster response times and lower token usage.

Ministral 3 3B sits inside the Ministral 3 3B family alongside Ministral 3 3B (Reasoning).

Its strongest category is Multilingual (#109), while its weakest is Knowledge (#121). This performance profile makes it a well-rounded choice across a range of tasks.

Creator

Mistral

Source Type

Open Weight

Reasoning

Non-Reasoning

Context Window

128K

Overall Score

27#121 of 123

Arena Elo

1037

Family & Lineage

Family

Ministral 3 3B

Base entry

Knowledge Benchmarks

MMLU
24
GPQA
23
SuperGPQA
21
OpenBookQA
19
MMLU-Pro
48
HLE
1
FrontierScience
28

Coding Benchmarks

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

Mathematics Benchmarks

AIME 2023
23
AIME 2024
25
AIME 2025
24
HMMT Feb 2023
19
HMMT Feb 2024
21
HMMT Feb 2025
20
BRUMO 2025
22
MATH-500
53

Reasoning Benchmarks

SimpleQA
22
MuSR
20
BBH
57
LongBench v2
32
MRCRv2
35

Agentic Benchmarks

Terminal-Bench 2.0
19
BrowseComp
33
OSWorld-Verified
20

Multimodal & Grounded Benchmarks

MMMU-Pro
25
OfficeQA Pro
37

Instruction Following Benchmarks

IFEval
67

Multilingual Benchmarks

MGSM
61
MMLU-ProX
59

Frequently Asked Questions

How does Ministral 3 3B perform overall in AI benchmarks?

Ministral 3 3B ranks #121 out of 123 models with an overall score of 27. It is created by Mistral and features a 128K context window.

Is Ministral 3 3B good for knowledge and understanding?

Ministral 3 3B ranks #121 out of 123 models in knowledge and understanding benchmarks with an average score of 24.5. There are stronger options in this category.

Is Ministral 3 3B good for coding and programming?

Ministral 3 3B ranks #121 out of 123 models in coding and programming benchmarks with an average score of 6.2. There are stronger options in this category.

Is Ministral 3 3B good for mathematics?

Ministral 3 3B ranks #114 out of 123 models in mathematics benchmarks with an average score of 36. There are stronger options in this category.

Is Ministral 3 3B good for reasoning and logic?

Ministral 3 3B ranks #121 out of 123 models in reasoning and logic benchmarks with an average score of 30.1. There are stronger options in this category.

Is Ministral 3 3B good for agentic tool use and computer tasks?

Ministral 3 3B ranks #121 out of 123 models in agentic tool use and computer tasks benchmarks with an average score of 22.9. There are stronger options in this category.

Is Ministral 3 3B good for multimodal and grounded tasks?

Ministral 3 3B ranks #121 out of 123 models in multimodal and grounded tasks benchmarks with an average score of 30.4. There are stronger options in this category.

Is Ministral 3 3B good for instruction following?

Ministral 3 3B ranks #113 out of 123 models in instruction following benchmarks with an average score of 67. There are stronger options in this category.

Is Ministral 3 3B good for multilingual tasks?

Ministral 3 3B ranks #109 out of 123 models in multilingual tasks benchmarks with an average score of 59.7. There are stronger options in this category.

Is Ministral 3 3B open source?

Yes, Ministral 3 3B is an open weight model created by Mistral, meaning it can be downloaded and run locally or fine-tuned for specific use cases.

Which sibling models are related to Ministral 3 3B?

Ministral 3 3B belongs to the Ministral 3 3B family. Related variants on BenchLM include Ministral 3 3B (Reasoning).

What is the context window size of Ministral 3 3B?

Ministral 3 3B has a context window of 128K, which determines how much text it can process in a single interaction.

Last updated: March 12, 2026

Weekly LLM Updates

New model releases, benchmark scores, and leaderboard changes. Every Friday.

Free. Your signup is stored with a derived country code for compliance routing.