Open WebUI

Open WebUI is a self-hosted web interface for chatting with AI models, like a ChatGPT-style app that runs on your own machine or server. It's one of the most-starred AI projects on GitHub.

Key features

• Any model: connect local models through Ollama, or cloud models through OpenAI, Anthropic or any OpenAI-compatible API (for example OpenRouter).
• Documents and retrieval (RAG): chat with your own files.
• Extensions: add web search, tools and Python "functions" and pipelines.
• Voice and vision: includes realtime voice calls (added in v0.12).
• Multi-user: user groups and permissions. The enterprise offering adds SSO, RBAC, audit logs and air-gapped deployment.
• Community library: shared prompts, models, tools and functions on openwebui.com.

Getting started

Install with pip install open-webui or Docker; no account with Open WebUI is needed to run it. The software is free, but you pay for whatever servers and model APIs you use. The code is on GitHub under the Open WebUI License, a BSD-style license with an added clause that forbids removing Open WebUI branding in larger deployments (above 50 users) unless an exception applies.

Good for

Individuals and teams who want one private chat front end for local and cloud models. For a desktop app instead of a web server, see Cherry Studio or LM Studio.