Instructions to use Open4bits/Ministral-3-3B-Base-2512-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Open4bits/Ministral-3-3B-Base-2512-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Open4bits/Ministral-3-3B-Base-2512-gguf",
	filename="ministral-3-3b-base-2512-q4_0.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Open4bits/Ministral-3-3B-Base-2512-gguf with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Open4bits/Ministral-3-3B-Base-2512-gguf:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Open4bits/Ministral-3-3B-Base-2512-gguf:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Open4bits/Ministral-3-3B-Base-2512-gguf:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Open4bits/Ministral-3-3B-Base-2512-gguf:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf Open4bits/Ministral-3-3B-Base-2512-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Open4bits/Ministral-3-3B-Base-2512-gguf:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf Open4bits/Ministral-3-3B-Base-2512-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Open4bits/Ministral-3-3B-Base-2512-gguf:Q4_K_M

Use Docker

docker model run hf.co/Open4bits/Ministral-3-3B-Base-2512-gguf:Q4_K_M

LM Studio
Jan

vLLM

How to use Open4bits/Ministral-3-3B-Base-2512-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Install mistral-common:
pip install --upgrade mistral-common
# Start the vLLM server:
vllm serve "Open4bits/Ministral-3-3B-Base-2512-gguf" --tokenizer_mode mistral --config_format mistral --load_format mistral --tool-call-parser mistral --enable-auto-tool-choice
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Open4bits/Ministral-3-3B-Base-2512-gguf",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Open4bits/Ministral-3-3B-Base-2512-gguf:Q4_K_M

Ollama
How to use Open4bits/Ministral-3-3B-Base-2512-gguf with Ollama:
```
ollama run hf.co/Open4bits/Ministral-3-3B-Base-2512-gguf:Q4_K_M
```

Unsloth Studio

How to use Open4bits/Ministral-3-3B-Base-2512-gguf with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Open4bits/Ministral-3-3B-Base-2512-gguf to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Open4bits/Ministral-3-3B-Base-2512-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Open4bits/Ministral-3-3B-Base-2512-gguf to start chatting

Docker Model Runner
How to use Open4bits/Ministral-3-3B-Base-2512-gguf with Docker Model Runner:
```
docker model run hf.co/Open4bits/Ministral-3-3B-Base-2512-gguf:Q4_K_M
```

Lemonade

How to use Open4bits/Ministral-3-3B-Base-2512-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Open4bits/Ministral-3-3B-Base-2512-gguf:Q4_K_M

Run and chat with the model

lemonade run user.Ministral-3-3B-Base-2512-gguf-Q4_K_M

List all available models

lemonade list

Open4bits / Ministral-3-3B-Base-2512-GGUF

This repository provides the Ministral 3 3B Base model converted to GGUF format, published by Open4bits to enable efficient local inference with reduced memory usage and broad CPU compatibility.

The underlying Ministral 3 model and architecture are developed and owned by Ministral. This repository contains only a quantized GGUF conversion of the original model weights.

The model is designed for lightweight, high-performance text generation and instruction-following tasks, making it well suited for local and resource-constrained environments.

Model Overview

Ministral 3 is a next-generation transformer-based large language model developed for strong generalization and robust natural language understanding. This release uses the 3B parameter Base variant, optimized for general-purpose text generation, reasoning, and instruction compliance.

The GGUF format enables broad compatibility with popular local inference engines and efficient CPU-based runtimes.

Model Details

Architecture: Ministral 3 Base
Parameters: ~3 billion
Format: GGUF (quantized)
Task: Text generation, instruction following
Weight tying: Preserved
Compatibility: GGUF-compatible inference runtimes (CPU-focused)

Compared to larger models in the same family, this variant offers a favorable balance of performance and resource efficiency.

Intended Use

This model is intended for:

Local text generation and conversational applications
CPU-based or low-resource deployments
Research, experimentation, and prototyping
Self-hosted or offline AI systems

Limitations

Reduced performance compared to larger or non-quantized variants
Output quality depends on prompt engineering and inference settings
Not specifically tuned for domain-specific or specialized tasks

License

This model is released under the original licensing terms of the base Ministral 3 model. Users must comply with the licensing conditions defined by the original model creators.

Support

If you find this model useful, please consider supporting the project. Your support enables Open4bits to continue releasing and maintaining high-quality, efficient open models for the community.

Downloads last month: 223

GGUF

Model size

3B params

Architecture

llama

Hardware compatibility

4-bit

5-bit

6-bit

8-bit

Inference Providers NEW

Image-Text-to-Text

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Open4bits/Ministral-3-3B-Base-2512-gguf

Base model

mistralai/Ministral-3-3B-Base-2512

Quantized

(31)

this model

Collection including Open4bits/Ministral-3-3B-Base-2512-gguf

GGUF

Collection

28 items • Updated Apr 27 • 1