Open WebUI: Self-hosted AI chat interface
Extensible multi-LLM chat platform with RAG pipeline.
Learn more about Open WebUI
Open WebUI is a self-hosted web application that provides a chat interface for multiple LLM backends and services. It operates as a containerized application deployable via Docker or Kubernetes, supporting both CPU and GPU execution through tagged image variants. The platform includes a built-in RAG engine for document processing, integrates with external APIs for web search, and offers administrative controls for user management and permissions. Common deployment scenarios include local Ollama instances, cloud-hosted API services, and enterprise environments with identity provider integration.
Unified Multi-Backend Interface
Connects to Ollama, OpenAI, and OpenAI-compatible endpoints through configurable API URLs. Switch between local and remote LLM sources without changing the chat interface or redeploying.
Built-in RAG Pipeline
Processes local documents and web search results directly within the application. Injects context into chat through command syntax without external vector databases or API dependencies.
SCIM 2.0 Provisioning
Automates user and group synchronization with enterprise identity providers like Okta and Azure AD. Provides role-based access control for multi-tenant deployments without manual account management.
from open_webui.tools import Tool
class WeatherTool(Tool):
def __init__(self):
self.name = "weather"
self.description = "Get current weather for a location"
async def run(self, location: str) -> dict:
# Tool logic to fetch weather data
return {"location": location, "temp": 72, "condition": "sunny"}Fixes workspace prompts editor, improves local Whisper STT support, and optimizes evaluations page performance.
- –Users can now create and save prompts in the workspace prompts editor without encountering errors
- –Users can now use local Whisper for speech-to-text when STT_ENGINE is left empty (the default for local mode)
- –The Evaluations page now loads faster by eliminating duplicate API calls to the leaderboard and feedbacks endpoints
- –Fixed missing Settings tab i18n label keys
Improved reliability for low-spec and SQLite deployments by disabling database session sharing by default.
- –Improved reliability for low-spec and SQLite deployments. Fixed page timeouts by disabling database session sharing by default
- –Users can re-enable via 'DATABASE_ENABLE_SESSION_SHARING=true' if needed
Introduces native function calling with built-in tools for multi-step tasks combining web research, knowledge bases, and memory.
- –Users with models that support interleaved thinking now get more refined results from multi-step workflows
- –When models invoke web search, search results appear as clickable citations in real-time for full source verification
- –Users can selectively disable specific built-in tools (timestamps, memory, chat history, notes, web search, knowledge bases) per model
- –Pending tool calls are now displayed during response generation, so users know which tools are being invoked
- –Administrators can now limit the number of files that can be uploaded to folders using the FOLDER_MAX_FILE_COUNT setting
See how people are using Open WebUI
Related Repositories
Discover similar tools and frameworks used by developers
Goose
LLM-powered agent automating local software engineering workflows.
Pi Mono
Monorepo providing AI agent development tools, unified LLM API, and deployment management for multiple providers.
InvokeAI
Node-based workflow interface for local Stable Diffusion deployment.
Wan2.1
Diffusion transformer models for text and image-to-video generation.
OpenCV
Cross-platform C++ library for real-time computer vision algorithms.