Open WebUI
Chat with Ollama models like DeepSeek-R1 and LLama, or use an OpenAI API key
Screenshots



About
Open WebUI lets you chat with advanced AI models running locally on your own device or connect to online models using an API key. **Getting Started with Local AI Models:** 🦙 Install Ollama: Start by installing the Ollama app from the Umbrel App Store. Ollama enables you to download and run large language models like Llama 3 and DeepSeek-R1 directly on your device. ⬇️ Download a Model: In the Open WebUI app, type the name of the model you want in the search bar and click "Pull from Ollama.com." A full list of models is available at https://ollama.com/. 🤖 Example - Running DeepSeek-R1 1.5B: To use the DeepSeek-R1 model with 1.5 billion parameters, type deepseek-r1:1.5b in the search bar and start the download. Once it's ready, you can chat with the model directly in Open WebUI. ⚠️ Warning: Before running a model, make sure your device has enough free RAM to support it. Attempting to run a model that exceeds your available memory could cause your device to crash or become unresponsive. Always check the model requirements before downloading or starting it.
Release Notes
This release introduces new features and improvements: - Enhanced communication with new Group and Direct Message channels, including unread counts, message pinning, and user status indicators - Improved security with sign-in rate limiting to protect against brute force attacks - Administrators can now globally disable the folders feature and control user-level folder permissions - Refined model editor interface and rebuilt knowledge base file management for better performance - Added Azure Document Intelligence model selection - General performance, stability, and security improvements - Expanded and improved translations Key bug fixes include: - Resolved issues with tool call responses and web search domain filtering - Corrected how custom models inherit connection types and improved active user count accuracy - Fixed default group assignment for all user registration methods - Addressed model list filtering, user profile image display, and Redis TLS connection failures - Resolved web page attachment issues and ensured model allowlist persistence in multi-worker setups - Prevented chat history infinite loading and improved database migration robustness - Corrected OAuth group assignment and knowledge base file access inconsistencies - Fixed embedding model auto-update functionality and reduced API response payload sizes - Eliminated redundant API calls on admin and workspace pages - Corrected button behavior, Ollama model deletion, and image generation in temporary chats - Fixed audit logging failures and dark mode styling - Resolved Milvus vector database query filtering issues Notable changes: - Channel creation is now restricted to administrators only - The active user count indicator was removed from the sidebar - The user table was restructured for API keys and OAuth data storage - The knowledge base API was restructured to support the new file relationship model Full release notes are found at https://github.com/open-webui/open-webui/releases