This repository contains a collection of extensions for Open WebUI, enhancing its functionality and user experience.
You can find the latest version of each extension on openwebui.com.
Automatically identifies and stores valuable information from chats as Memories in Open WebUI. This extension analyzes conversations and extracts key details to remember for future interactions.
- Automatically captures relevant information from user messages
- Consolidates similar or overlapping memories to avoid duplication
- Updates conflicting information with the most recent data
- Optionally saves assistant responses as memories
- Configurable memory processing parameters
- OpenAI API URL: Configure a compatible endpoint for memory processing
- Model: Select which model to use for memory identification
- API Key: Set your API key for the chosen endpoint
- Related Memories: Control how many previous memories to consider when updating
- Memory Distance: Adjust the threshold for determining related memories
- Save Assistant Responses: Option to automatically save assistant messages as memories
- Legacy Mode: Use simpler memory processing for older setups
- Messages to Consider: Set how many recent messages to analyze for memory extraction
- Navigate to the Open WebUI Extensions page
- Click on the "Get" button to automatically install the extension
- Configure the extension through the Open WebUI settings interface
- Clone this repository:
git clone https://github.com/davixk/open-webui-extensions.git
- Copy the desired extension(s) to your Open WebUI functions directory:
cp -r functions/auto_memory_function.py /path/to/open-webui/functions/
- Restart Open WebUI to load the new extensions.
Each extension can be configured through the Open WebUI settings interface under the "Extensions" section. Enable the desired extensions and adjust their settings as needed.
To create new extensions for Open WebUI:
- Use the existing extensions as templates
- Follow the function structure with
inlet
andoutlet
methods - Create appropriate configuration models using Pydantic
- Add comprehensive documentation in the extension's docstring
Contributions are welcome! Please feel free to submit a Pull Request.
[Specify your license information here]
For questions, issues, or feature requests, please open an issue on GitHub.