Running AI models in a Docker container ensures portability, isolation, and efficient execution. If you want to deploy DeepSeek-R1 8B inside a Docker container using Ollama, this guide will walk you through the step-by-step process.
Why Use Docker for DeepSeek-R1 8B?
- Portability – Run the model on any system with Docker installed.
- Isolation – Keeps dependencies contained within the container.
- Scalability – Easily deploy across multiple environments.
Step-by-Step Deployment Guide
1. Run the Ollama Docker Container
First, start the Ollama container by running the following command:
docker run -d -v ollama:/root/.ollama -p 11434:11434 --name ollama ollama/ollamaThis will:
- Start the Ollama container in detached mode (
-d). - Mount the Ollama volume for persistent storage.
- Expose the API on port
11434.
2. Log into the Ollama Container
To access the running Ollama container, use:
docker exec -it ollama /bin/bashThis will allow you to interact with the container’s shell.
3. Pull the DeepSeek-R1 8B Model
Once inside the container, pull the required DeepSeek-R1 8B model:
ollama pull deepseek-r1:8bThis command fetches all necessary dependencies.
4. Run the Model
Now, run the downloaded model:
ollama run deepseek-r1:8bThis will launch the model in interactive mode, allowing you to enter prompts.
Example:
>>> Hello!5. Run the Ollama Web GUI Container
To enable a web-based interface, run the Ollama Web UI container. Replace <YOUR-IP> with your system’s local IP:
docker run -d -p 3000:8080 -e OLLAMA_BASE_URL=http://<YOUR-IP>:11434 \
-v open-webui:/app/backend/data --name open-webui --restart always \
ghcr.io/open-webui/open-webui:main6. Access the Web Interface
Now, open your browser and go to:
http://<YOUR-IP>:3000You can now interact with the DeepSeek model using a user-friendly web interface.
Conclusion
Deploying DeepSeek-R1 8B with Docker and Ollama provides a robust and efficient way to run AI models in a controlled environment. With Docker, you gain portability and isolation, while Ollama simplifies the model execution. Whether through command-line interaction or a web-based UI, this setup ensures seamless AI deployment.