Open WebUI

What Open WebUI is

Open WebUI is a free, self-hosted interface for using local and cloud AI models, built to run entirely offline. It connects to Ollama and OpenAI-compatible APIs, and installation options include Docker, pip, uv and desktop setup; Docker or Python is recommended for production deployments. Listed capabilities include voice, vision, retrieval, generation, search and Python extensions, along with web search, file uploads and image generation. Its plugins include Filters, Actions, Pipes, Tools and Skills, with connections through MCP, MCPO and OpenAPI tool servers. Knowledge retrieval includes local RAG with nine vector databases, multiple content-extraction engines, hybrid search, reranking and full-context mode. Organizations can deploy it on-premises, in a private cloud or in an air-gapped environment. Chats, files, knowledge bases, embeddings, users, permissions and logs are stored in operator-configured storage, though configured external providers, tools, model endpoints or integrations can route data externally. Security controls include LDAP and Active Directory, OIDC and OAuth 2.0 SSO, role and model permissions, and workspace isolation. The software is available across web, mobile, desktop and self-hosted platforms, with app integrations and a team workspace.