Instructions to use SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M # Run inference directly in the terminal: llama cli -hf SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M # Run inference directly in the terminal: llama cli -hf SandLogicTechnologies/Ministral-3-8B-Instruct-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 SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M # Run inference directly in the terminal: ./llama-cli -hf SandLogicTechnologies/Ministral-3-8B-Instruct-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 SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M
Use Docker
docker model run hf.co/SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M
- LM Studio
- Jan
- vLLM
How to use SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M
- Ollama
How to use SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF with Ollama:
ollama run hf.co/SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M
- Unsloth Studio
How to use SandLogicTechnologies/Ministral-3-8B-Instruct-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 SandLogicTechnologies/Ministral-3-8B-Instruct-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 SandLogicTechnologies/Ministral-3-8B-Instruct-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 SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF to start chatting
- Pi
How to use SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF with Docker Model Runner:
docker model run hf.co/SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M
- Lemonade
How to use SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M
Run and chat with the model
lemonade run user.Ministral-3-8B-Instruct-2512-GGUF-Q4_K_M
List all available models
lemonade list
- Hermes Agent
How to use SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:8080/v1 hermes config set model.default SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M
Configure OpenClaw
# Install OpenClaw: npm install -g openclaw@latest # Register the local server and set it as the default model: openclaw onboard --non-interactive --mode local \ --auth-choice custom-api-key \ --custom-base-url http://127.0.0.1:8080/v1 \ --custom-model-id "SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M" \ --custom-provider-id llama-cpp \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
Ministral-3-8B-Instruct (Vision-Language & vLLM Compatible)
Ministral-3-8B-Instruct is a vision-aware, instruction-tuned multimodal language model developed by Mistral AI. It combines textual and visual understanding with strong reasoning capabilities and reliable instruction adherence.
This repository provides Q4_K_M and Q5_K_M quantized variants of the model, optimized for efficient local inference. These quantized formats reduce memory usage and improve inference performance while retaining support for vision-language interaction with both text and image inputs.
Model Overview
- Model Name: Ministral-3-8B-Instruct
- Base Model: mistralai/Ministral-3-8B-Instruct-2512
- Architecture: Transformer-based multimodal model
- Parameter Count: 8 Billion
- Contexts Supported: Text & Images
- Developer: Mistral AI
- License: Apache 2.0
Quantization Formats
Q4_K_M
- Approx.
71% size reduction (4.84 GB) - Substantial reduction in model size
- Designed for low-memory environments
- Faster inference on CPU-based systems
- Suitable for lightweight and edge use cases
Q5_K_M
- Approx.
66% size reduction (5.64 GB) - Higher precision than Q4 variants
- Improved response consistency and reasoning depth
- Recommended for balanced performance and quality
Vision-Language Capabilities
Ministral-3-8B-Instruct supports multimodal inputs, allowing users to provide both text and images within the same prompt. This enables applications such as:
- Image captioning and explanation
- Visual question answering
- Instruction following grounded in vision
- Contextual multimodal analysis
The model processes textual and visual information jointly, producing coherent responses that factor in both modalities.
Training Background
The base model was pretrained on a large mixture of text and visual data, followed by instruction tuning that emphasizes reliable multimodal reasoning and instruction compliance.
Pretraining
- Large-scale multimodal pretraining
- Joint text-image representation learning
- Optimization for robust, coherent generation
Instruction Tuning
- Fine-tuned with multimodal instruction datasets
- Trained for clarity, task adherence, and visual reasoning
- Enhanced for conversational quality across modalities
Key Capabilities
Multimodal Input Understanding
Incorporates image content and text together to produce aligned responses.Instruction Compliance
Follows detailed user directives, including ones involving visual context.Reasoning & Analysis
Supports step-by-step explanation and problem solving, integrating visual evidence.Conversational Dialogue
Maintains fluid dialogue across mixed text-image interaction.Efficient vLLM Serving
Works well with vLLM inference for scalable deployment.
Usage Examples
LLama.cpp Usage
/llama-cli \
-m SandlogicTechnologies\ministral-3-8b-instruct_Q4_K_M.gguf \
--image ./example.png \
-p "Explain what is happening in this image."
Recommended Applications
Multimodal Assistants Build systems that understand and respond to both images and text.
Visual QA Tools Create applications that answer questions grounded in image context.
Content Understanding Use for summarizing or reasoning over documents with associated images.
Conversational AI Serve rich, multimodal dialogues in high-throughput environments.
Acknowledgments
This repository is based on the Ministral-3-8B-Instruct model, developed by Mistral AI.
Thanks to:
- The Mistral AI team for releasing multimodal capabilities
- The
llama.cppcommunity for enabling efficient GGUF inference
Contact
For questions, feedback, or support, please reach out at support@sandlogic.com or visit https://www.sandlogic.com/
- Downloads last month
- 41
4-bit
5-bit
Model tree for SandLogicTechnologies/Ministral-3-8B-Instruct-2512-GGUF
Base model
mistralai/Ministral-3-8B-Base-2512