Skip to main content

Ministral 3 8B

MistralMistralOpen WeightMistral Research License

Description

The Ministral-8B-Instruct-2410 is an instruct fine-tuned model for local intelligence, on-device computing, and at-the-edge use cases, significantly outperforming existing models of similar size.

Release Date
2025-12-02
Parameters
8.0B
Context Length
262K
Modalities
image, text

Capability Radar

28
general
18
coding
34
reasoning
29
scienceest.
0
agents
10
multimodal

Science uses a reasoning proxy when dedicated science benchmarks are unavailable.

Rankings

Domain#RankScoreSource
Code Ranking329
21.0
AA
General Ranking354
31.0
AA
Math Reasoning255
32.0
AA
Science365
29.0
AA

Benchmark Scores (LLM Stats)

Code

HumanEval34.8%SR

Communication

MT-Bench0.83 / 100SR

Creativity

Arena Hard70.9%SR

Finance

MMLU65.0%SR
French MMLU57.5%SR

General

ARC-C71.9%SR
MBPP pass@170.0%SR
TriviaQA65.5%SR
AGIEval48.3%SR

Language

Winogrande75.3%SR

Math

MATH54.5%SR

AA Evaluation Indices

Math Index
31.7
Intelligence Index
14.8
Coding Index
10.0
Mmlu Pro
0.6
Gpqa
0.5
Aime 25
0.3
Livecodebench
0.3
Ifbench
0.3
Tau2
0.3
Lcr
0.2
Scicode
0.2
Terminalbench Hard
0.0
Hle
0.0

LLM Stats Category Scores

Communication
80
Creativity
80
Roleplay
80
Writing
70
Finance
70
General
70
Healthcare
70
Language
70
Legal
60
Math
60
Reasoning
60
Code
30

Pricing

Input Price$0.15 / 1M tokens
Output Price$0.15 / 1M tokens
Blended Price (3:1)$0.15 / 1M tokens

Speed

Tokens/sec167.2 tokens/s
Time to First Token0.35s
Time to Answer0.35s

Available Providers

(LS internal units)

No provider data available

External Sources