models/mistral/ministral-3b-2512
M
Mistral·active

Ministral 3 3B 2512

Mistral's efficiency model. Context window: 131K tokens.

Overall score
3.31
/5.00 · ranked #77
Input
$0.100
per 1M tokens
Output
$0.100
per 1M tokens
Context
131K
tokens
Blended
$0.100
3:1 out:in ratio

Price drops, new benchmarks, model updates. Stay current on Ministral 3 3B 2512.

One email per change. Unsubscribe anytime.

modelpicker.aipowered by live benchmark data

Scores by test

Methodology →
Structured Output
4.0
Strategic Analysis
2.0
Constrained Rewriting
5.0
Creative Problem Solving
3.0
Tool Calling
4.0
Faithfulness
5.0
Classification
4.0
Long Context
4.0
Safety Calibration
1.0
Persona Consistency
4.0
Agentic Planning
3.0
Multilingual
4.0
Tabular Data

What you need to know

Ministral 3 3B 2512 is a high-precision, low-cost model optimized for reliability and strict adherence to instructions. Its primary differentiator is its perfect score in faithfulness and constrained rewriting, making it highly effective for tasks where hallucination must be minimized and output formats must be strictly maintained.

At $0.100 per million tokens for both input and output, the model is priced aggressively for its capabilities. While it ranks lower overall (#58 of 71), its strength in structured output and tool calling suggests it functions better as a specialized utility model than a general-purpose reasoning engine.

The model struggles with high-level cognitive tasks, showing significant weaknesses in strategic analysis and creative problem solving. Additionally, its safety calibration is minimal, meaning developers must implement their own robust filtering layers if the model is exposed to end-users.

Use this model for high-volume data transformation, structured rewriting, and basic tool-calling agents where cost efficiency and factual accuracy are priorities. Skip this model for complex strategic planning, creative generation, or applications requiring built-in safety guardrails.

Strengths — Top 3

Constrained Rewriting5.0/5.0
Faithfulness5.0/5.0
Structured Output4.0/5.0

Relative weaknesses — Bottom 3

Safety Calibration1.0/5.0
Strategic Analysis2.0/5.0
Creative Problem Solving3.0/5.0

Similar models

MMistral Small 3.2 24B$0.1693.00MCodestral 2508$0.7503.23MDevstral Small 1.1$0.2502.85QQwen: Qwen3 30B A3B Instruct 2507$0.2473.62