Open WebUI

Overview

Open WebUI is an AI platform you install and run on your own server, and it is designed to work offline. It supports Ollama and OpenAI-compatible APIs, so you can use local and cloud models together.

It is installed with pip, uv, Docker or Kubernetes. You can build specialized agents by bundling instructions, tools and knowledge with a base model, and connect external services through Filters, Actions, Pipes, Tools and Skills plugins and MCP. It also has Open Terminal, which gives agents a terminal and file system, as well as notes, team channels and memory across conversations.

It suits individuals and organizations that want to keep data in their own environment while using multiple models. Administrators can set fine-grained permissions by role and group.

What it is used for

• Chat with Ollama's local models and cloud models in one interface.
• Build specialized internal agents that bundle instructions, tools and knowledge.
• Connect external services and data through MCP and plugins.
• Refine documents together with colleagues and AI models in team channels.

Pricing and license

Enterprise plan sold separately. See the official site for licensing

Getting started

1. Install with pip install open-webui or run the Docker image.
2. Connect an Ollama or OpenAI-compatible API address.
3. Choose a model and start chatting, adding tools and plugins if needed.

Links

Official site ↗Repository ↗

Related terms and companies

Local LLM MCP Sovereign AI Agent memory Skills RAG Ollama

Sources

• open-webui/open-webui README
• Open WebUI: Self-Hosted AI Platform

Compiled from public sources (official pages, READMEs). Pricing and features change, so check the official site.