Imagine a ChatGPT-like interface — conversation threads, model switching, file uploads — but everything runs on your machine. No data leaves your network. That’s exactly what Open WebUI gives you when paired with Ollama.
Open WebUI (formerly Ollama WebUI) is the most popular web frontend for Ollama, with over 50K GitHub stars. It takes about two minutes to set up and turns your local LLM setup into something your whole team can use.
What You Need
Before starting, make sure you have:
- Ollama installed and running — if you haven’t set it up yet, follow our complete Ollama guide
- Docker installed — the easiest way to run Open WebUI (alternatively, you can use pip — covered below)
- At least one model pulled in Ollama (e.g.,
ollama pull llama3)
That’s it. No GPU required for Open WebUI itself — it’s just the frontend. The heavy lifting happens in Ollama.
Step 1: Start Ollama
Make sure Ollama is running in the background. Open a terminal and run:
ollama serveIf you installed Ollama as a system service, it’s likely already running. You can verify by visiting http://localhost:11434 in your browser — you should see “Ollama is running.”
Pull a model if you haven’t already:
ollama pull llama3
ollama pull mistralStep 2: Run Open WebUI with Docker
This is a single command. Copy, paste, done:
docker run -d -p 3000:8080 \
--add-host=host.docker.internal:host-gateway \
-v open-webui:/app/backend/data \
--name open-webui \
ghcr.io/open-webui/open-webui:mainHere’s what each flag does:
-d— runs the container in the background-p 3000:8080— maps port 3000 on your machine to the container’s port 8080--add-host=host.docker.internal:host-gateway— lets the container reach Ollama running on your host machine-v open-webui:/app/backend/data— persists your data (conversations, settings, users) across container restarts--name open-webui— gives the container a recognizable name
Docker will pull the image and start the container. Wait about 30 seconds for it to initialize.
Step 3: Create Your Account
Open your browser and go to:
http://localhost:3000You’ll see a sign-up screen. Create your account — the first user to register automatically becomes the admin. This is important: the admin can manage users, configure connections, and control which models are available.
Once logged in, Open WebUI should automatically detect your local Ollama instance and list your available models in the model selector dropdown.
Step 4: Start Chatting
Select a model from the dropdown at the top of the chat window and start typing. It works exactly like ChatGPT:
- Type your message, hit Enter
- Responses stream in real-time
- Conversations are saved automatically in the sidebar
- Switch models mid-conversation if you want to compare outputs
Everything stays local. Your prompts, responses, and conversation history never leave your machine.
Key Features Worth Exploring
Open WebUI is far more than a basic chat window. Here are the features that make it stand out:
RAG — Document Upload and Chat
Upload PDFs, text files, or other documents directly into a conversation. Open WebUI will chunk and index them, letting you ask questions about the content using your local model. No external APIs involved. For a deeper dive into building RAG pipelines, see our local RAG pipeline guide.
Multi-User Support
As admin, you can invite other users. Each person gets their own conversation history and settings. This makes Open WebUI perfect for teams — set up one server with Ollama and a capable self-hosted model, then give everyone on your team access through the web UI. All data stays on your infrastructure, which is a big win for GDPR compliance.
Model Switching
Switch between any model you’ve pulled in Ollama with a single click. Running llama3 for general chat and codellama for coding tasks? Just pick the one you need from the dropdown. You can even connect to the OpenAI API as a fallback if you want access to cloud models alongside your local ones.
Other Notable Features
- Web search — let your model search the web for up-to-date information
- Image generation — connect to local image generation backends
- Voice input — speak your prompts instead of typing
- Custom system prompts — set per-conversation or global instructions
- Markdown rendering — responses render with proper formatting, code blocks, and syntax highlighting
Running Without Docker (pip Install)
If you prefer not to use Docker, you can install Open WebUI directly with pip:
pip install open-webui
open-webui serveThis starts the server on http://localhost:8080 by default. It expects Ollama to be reachable at http://localhost:11434.
The Docker approach is recommended for most users because it isolates dependencies and makes updates simple (docker pull ghcr.io/open-webui/open-webui:main and restart the container). But the pip install works fine for quick local testing.
Connecting to a Remote Ollama Instance
If Ollama runs on a different machine (say, a GPU server on your network), you can point Open WebUI to it using the OLLAMA_BASE_URL environment variable:
docker run -d -p 3000:8080 \
-e OLLAMA_BASE_URL=http://192.168.1.100:11434 \
-v open-webui:/app/backend/data \
--name open-webui \
ghcr.io/open-webui/open-webui:mainReplace 192.168.1.100 with the IP of your Ollama server. Make sure Ollama is configured to listen on 0.0.0.0 (set OLLAMA_HOST=0.0.0.0 on the server) so it accepts connections from other machines.
Note that when connecting to a remote Ollama instance, you don’t need the --add-host flag since you’re not routing through the Docker host.
Wrapping Up
Open WebUI turns Ollama from a command-line tool into a full-featured, team-ready AI interface. One Docker command, a browser tab, and you’ve got a private ChatGPT alternative running entirely on your hardware.
The combination is hard to beat: Ollama handles model management and inference, Open WebUI handles the user experience. Add RAG, multi-user support, and conversation history on top, and you’ve got a setup that rivals commercial offerings — without sending a single byte to the cloud.