Open WebUI
Open WebUI is a self-hosted AI interface that connects local Ollama models and OpenAI-compatible APIs in a feature-rich workspace.
Open WebUI is a self-hosted AI platform designed to run entirely offline. It provides a single interface for local Ollama models and OpenAI-compatible APIs, so different providers can be used from the same workspace. The README describes installation through pip, uv, Docker, or Kubernetes, with tagged images for Ollama and CUDA deployments. It includes role-based access control and groups, plugins and tools via Filters, Actions, Pipes, Skills, MCP, MCPO, and OpenAPI servers, plus custom model instructions, knowledge, and agents. The workspace also includes Notes, shared Channels, persistent memory, live workflow display, calendars and AI scheduling, recurring automations, responsive PWA behavior, Markdown and LaTeX, voice and video calls, local RAG with multiple vector databases, web search, image generation and editing, and model evaluation features. The project supports SQLite or PostgreSQL and can store files locally or in S3-compatible and cloud storage. To try it, follow the official documentation and choose a deployment path that matches your setup. Open WebUI is intended for people who want an extensible AI interface that can connect to both local models and cloud-compatible model endpoints.