open-webui

Working with large language models often means juggling separate tools for local models, cloud providers and document workflows. Open WebUI is a self-hosted AI platform that brings them together in one browser-based interface. It supports Ollama and OpenAI-compatible APIs, so you can connect services such as LMStudio, OpenRouter, vLLM and Mistral alongside local models and switch between them freely. Beyond chat, the platform includes retrieval-augmented generation over documents, web search and browsing, image generation, multi-model conversations, notes, channels, and plugins built from filters, tools and MCP or OpenAPI tool servers. Administrators manage access with granular roles and user groups, and enterprise deployments can use LDAP, SSO and SCIM provisioning. It is designed for individuals, teams and organizations that want a private, provider-agnostic AI workspace. It can be installed with pip or Docker, deployed on Kubernetes, and run entirely offline.