How to add an AI chat interface with Open WebUI
Updated 7/20/2026
Open WebUI gives you a modern chat interface, similar to ChatGPT, that connects to your local models run with Ollama.
Requirements
Ollama installed and working, see how to run local LLMs with Ollama, and Docker.
1. Start Open WebUI
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:main
2. Access the interface
Open http://SERVER_IP:3000 and create the first account, which becomes the administrator.
3. Useful features
- Multiple conversations, like in ChatGPT.
- Multiple users with separate accounts.
- Document upload for questions based on them (RAG), see RAG chatbot.
Securing it
Serve only over HTTPS through a reverse proxy and limit access with the firewall. You now have your own private ChatGPT, fully under your control.
Put this guide into practice on your own VPS
EU cloud servers, billed hourly, ready in minutes, with one-click installs for dozens of apps.
See pricingRelated tutorials
How to use OpenHands as an AI software engineer
OpenHands takes a task and writes, runs and debugs the code on its own. Here is how to use it safely.
Updated 7/20/2026AI & self-hostingHow to extract web data without code, using Maxun
Build robots that collect data from websites by clicking, without writing a line of code.
Updated 7/20/2026AI & self-hostingHow to build a customer chatbot with Dify
Use Dify to create an assistant that answers from your company documents, without writing code.
Updated 7/20/2026