Open WebUI

Open WebUI is an open-source Web App and Server that serves as a comprehensive front-end interface for interacting with large language models, designed to operate entirely offline or connected to cloud services. The software is deployed as a server and accessed via a web browser or as a Progressive Web App for a native experience on mobile and desktop devices.

The platform connects to various LLM runners, including Ollama and any OpenAI-compatible APIs. This allows users to integrate models from providers such as GroqCloud, Mistral, and OpenRouter, enabling a mix of local and remote providers within a single interface. It is used for managing AI conversations, building specialized agents with custom instructions, and implementing local document retrieval for knowledge management.

Key features

• Integration with Ollama and OpenAI-compatible APIs
• Local Retrieval Augmented Generation with support for nine vector databases
• Role-based access control for managing user groups and permissions
• Plugin system supporting Filters, Actions, Pipes, Tools, and Skills
• Built-in calendar and AI scheduling via function calling
• Voice and video call integration with multiple STT and TTS providers
• Web search integration through providers like SearXNG and Brave Search
• Image generation and editing via DALL-E, ComfyUI, and AUTOMATIC1111
• Support for LDAP, Active Directory, and OAuth SSO authentication

Open WebUI is built for both individual users and enterprise environments. It supports horizontal scalability using Redis for session management and offers production observability through OpenTelemetry. The architecture allows for flexible storage options, including SQLite, PostgreSQL, and S3-compatible cloud storage. It also integrates with the Model Context Protocol (MCP) and OpenAPI tool servers to extend AI capabilities. Users can create shared spaces called Channels for real-time team collaboration and use a dedicated Notes workspace for drafting content outside of standard chat threads.

This software positions itself as a comprehensive AI orchestration layer that prioritizes data control, privacy, and extensibility.