Open WebUI
A self-hosted AI chat interface for Ollama, OpenAI-compatible APIs, and cloud models, with offline use and team access controls.
Open WebUI gives individuals and teams a self-hosted place to chat with local LLMs and cloud models. It runs on your own computer or server, including through Docker, and can work entirely offline with local models.
It connects to Ollama and OpenAI-compatible APIs such as those from LM Studio, vLLM and OpenRouter. You can also use OpenAI and Anthropic models. This lets you keep local models and cloud services in one interface, with the choice of model determining where requests are processed.
The workspace goes beyond chat. You can use retrieval and search, work with voice and images, and give models access to tools. Model profiles can include instructions and knowledge for specialized agents. Python extensions and connections to MCP and OpenAPI tool servers provide ways to add capabilities, while a community library offers prompts and tools made by other users. Agents can work with files through a terminal and return the results in chat.
Notes provide a separate space for drafting material that you can bring into a conversation. Shared channels let team members and models work in threads, with reactions and pinned messages. For organizations managing access and data location, administrators can set roles and group permissions; the platform also supports SSO, audit logs, and on-premises or air-gapped deployment.
