← Back to Discover

Open WebUI

Self-hosted AI chat platform that connects to Ollama and OpenAI-compatible APIs, with roles and groups, plugins, retrieval over your documents, web search, voice, image generation and agent features.

General assistantsLocal models
Visit website ↗

What it does

Open WebUI is a self-hosted AI platform that you run on your own infrastructure and can operate fully offline. It gives you one interface for local models served through Ollama and for any OpenAI-compatible API, so you can mix local and cloud providers. Beyond chat, it includes administrator-defined roles and groups, a plugin system with MCP and OpenAPI tool servers, custom agents built from a base model plus instructions, tools and knowledge, retrieval over your documents backed by nine vector databases, web search through many providers, voice and video calls, image generation and editing, notes, shared channels, a calendar and scheduled prompts. It installs with pip, Docker, Kubernetes or Helm. The README says the code is under multiple licences, including the Open WebUI License, which requires keeping the Open WebUI branding. An Enterprise plan is available through its sales team, and some items, such as per-user isolated terminals, are marked as Enterprise.

Capabilities

  • Works with Ollama and any OpenAI-compatible API
  • Roles and user groups, with LDAP, SSO and SCIM options
  • Plugins, plus MCP and OpenAPI tool servers
  • Document retrieval with hybrid search and nine vector databases
  • Web search, image generation and editing, and voice or video calls
  • Custom agents built from instructions, tools and knowledge

When to use it

  • You want a private, self-hosted chat interface for local and cloud models
  • A team needs shared access with permissions and groups
  • You want document question-answering set up inside your own deployment