Imagine generating stunning AI images without worrying about data privacy, credit limits, or annoying content filters. With Docker Model Runner and Open WebUI, you can build a fully local image generation setup—think of it as your own private DALL-E, no cloud subscription needed. This guide walks you through the entire process, from prerequisites to generating your first image, all while keeping your data on your machine.
1. Understand the Big Picture: How It All Fits Together
Docker Model Runner acts as the control plane for local AI inference. It downloads models, manages the backend lifecycle, and exposes a 100% OpenAI-compatible API, including the /v1/images/generations endpoint. Open WebUI, a powerful chat interface, connects to this endpoint seamlessly, allowing you to generate images through a familiar chat UI—fully local, fully private. The magic happens in two commands: one to pull the model, another to launch the UI. No complex configuration, no cloud dependencies—just your hardware and your imagination.
2. Check Your System Requirements
Before diving in, ensure your machine is up to the task. You'll need Docker Desktop (macOS) or Docker Engine (Linux). For memory, aim for at least 8 GB of free RAM—more is better for larger models. A GPU is optional but highly recommended for speed; NVIDIA CUDA, Apple Silicon MPS, or even CPU fallback works. To verify your setup, run docker model version. If it returns without errors, you're ready to proceed. The entire process is designed to be lightweight, but a decent GPU will make generation nearly instant.
3. Pull an Image Generation Model
Docker Model Runner uses a compact packaging format called DDUF (Diffusers Unified Format) to distribute models via Docker Hub, just like any other OCI artifact. To get started, open a terminal and run: docker model pull stable-diffusion. This downloads the model locally. Confirm it's ready with docker model inspect stable-diffusion, which shows details like the model ID, tags, and the bundled components. Under the hood, the DDUF file packs a text encoder, VAE, UNet/DiT, and scheduler config into a single portable artifact. At runtime, Docker Model Runner unpacks it automatically—no manual extraction needed.
4. Launch Open WebUI with One Simple Command
This is where the magic happens. Docker Model Runner has a built-in launch command that automatically wires Open WebUI against your local inference endpoint. Simply run: docker model launch openwebui. That's it—no environment variables, no manual configuration. The command pulls the Open WebUI container, connects it to the local model API, and opens a browser tab ready for interaction. You can now access a full-featured chat interface that supports image generation alongside text, all running locally. The integration is seamless, making it feel like a cloud service without the cloud.
5. Generate Your First Image from the Chat Interface
Once Open WebUI is running, click into the chat window and type a prompt like "a dragon wearing a business suit" or "a serene mountain landscape at sunset." The model will process your request and return the generated image directly in the chat. Because everything is local, your prompts never leave your machine—no data logs, no credit deductions, no content filters overruling your creativity. You can experiment with different styles, adjust parameters, and generate multiple images instantly. The response time depends on your hardware, but even on a CPU you'll see results in minutes. For the best experience, use a GPU with at least 8 GB VRAM.
6. Explore Advanced Features and Customization
Open WebUI offers more than just image generation. You can switch between different models, adjust inference settings like steps and guidance scale, and even use negative prompts. If you want to try other models, just pull them with Docker Model Runner—similarly to the first step. Models are available from the official Docker Hub repository, and you can inspect available tags with docker model list. For power users, the setup supports multiple concurrent users and integrates with other AI tools. The local nature also allows you to modify models or fine-tune them, though that's beyond this basic walkthrough.
7. Enjoy the Benefits of Fully Private, No-Subscription AI
By running image generation locally, you gain complete control over your data. No third-party servers see your prompts, no subscription fees drain your wallet, and no arbitrary filters block your creative visions. Your machine does all the work, and only you see the results. This setup is ideal for privacy-conscious individuals, small teams, or anyone who wants to experiment freely with AI image generation without recurring costs. The investment is one-time hardware and a bit of disk space. With Docker Model Runner and Open WebUI, the power of cutting-edge AI is truly at your fingertips—no cloud required.
Now you have a complete, local image generation system. Start creating, sharing, and exploring—all from the comfort of your own machine. The future of private AI is here, and it's running on your desktop.