Deepseek-R1 Local Deployment with Docker & Open WebUI

This project provides a ready-to-use Docker Compose setup for running the DeepSeek-R1 language model locally, along with a simple web-based user interface.

Features

• Local LLM Inference: Run DeepSeek-R1 models on your own hardware using Ollama.
• Web UI: Interact with the model via a browser-based interface in the web/ directory.
• Easy Model Management: Pull and manage model versions with Docker commands.

Quick Start

1. Start the Ollama Service

   docker compose up -d ollama
   docker compose up -d web
   docker compose up -d open-webui

2. Pull the DeepSeek-R1 Model

   Choose your desired model size (e.g., 1.5b):

   docker compose exec ollama ollama pull deepseek-r1:1.5b

3. Access the Web UI

   • Custom UI http://localhost:6001
   • Open WebUI http://localhost:6002

   Open http://localhost:6002 in your browser to interact with the model.

4. Add new models

   docker compose exec ollama ollama pull deepseek-r1:7b
   docker compose exec ollama ollama pull deepseek-coder:1.3b

Project Structure

.
├── docker-compose.yml         # Docker Compose configuration for Ollama
├── ollama-models/             # Model storage (keys, manifests, blobs)
│   ├── id_ed25519
│   ├── id_ed25519.pub
│   └── models/
│       ├── blobs/             # Model weights and data
│       └── manifests/         # Model manifests
├── web/                       # Web UI files
│   ├── index.html
│   ├── ollama.js
│   ├── showdown.min.js
│   └── style.css
└── README.md                  # Project documentation

Notes

• Model files are stored in ollama-models/. You can add or remove models as needed.
• The web UI is static and communicates with the Ollama backend.
• For advanced configuration, edit docker-compose.yml.

Troubleshoting

• Check Ollama is runing on http://localhost:11434
• Custom Web UI is running on http://localhost:6001
• Open Web UI is running on http://localhost:6002

References

• Ollama Documentation
• DeepSeek-R1 Model Card
• https://dev.to/savvasstephnds/run-deepseek-locally-using-docker-2pdm
• https://platzi.com/blog/deepseek-r1-instalar-local/
• https://www.composerize.com/

---