Open WebUI is a self-hosted interface for running AI models on your own machine, in your cloud, or fully air-gapped. It connects to local models and to cloud providers such as Ollama, OpenAI and Anthropic, so one place holds your conversations, models and tools.
The platform is built around control. The page says your models, data and deployment stay where you choose, and that you can extend it with Python functions and tools. Prompts, models, tools and functions are shared in a community library at openwebui.com, where members can browse, install and contribute.
For teams, the page lists SSO, role-based access control, audit logs, data residency and on-premises or air-gapped deployment. It describes these as aimed at regulated industries as well as startups and larger enterprises. Sovereign AI, Open Terminal and a switch guide for running a pilot are linked from the site for organisations that want to plan a rollout.
Installation is offered through pip and Docker. The page gives the command pip install open-webui and says no account is required to start. The source is on GitHub under the open-webui project, and the page reports more than 460 million downloads and over 533,000 community members.
The page also lists a Discord server, a blog with release notes, and an enterprise option for organisations that want a supported deployment. Pricing for paid enterprise plans is not stated on the page, so the listing shows no starting price.
