Open WebUI
Self-hosted web interface for chatting with local and cloud AI models
Open WebUI is an extensible, self-hosted platform for chatting with AI models in the browser. It is for people who want a ChatGPT-style interface that connects to models running on their own machine through Ollama, or to OpenAI-compatible APIs.
The README says it is built to run entirely offline and can use local or cloud models. It has several Docker image variants, for CPU, CUDA graphics cards and a bundled Ollama, and a single docker run command starts it, with its data kept in a Docker volume on your own server.
Quick start
For a quick try only. See the official docs for data volumes, environment variables and a production setup.
ghcr.io/open-webui/open-webui:maindocker run -d \
--name open-webui \
-p 8080:8080 \
--restart unless-stopped \
ghcr.io/open-webui/open-webui:mainThen open http://localhost:8080