GraphRAG Web UI is a local knowledge base management and Q&A system based on Streamlit. It allows users to create, manage, delete, and query the knowledge base through an easy-to-use web interface.
GraphRAG Web UI is built using Microsoft's open-source library GraphRAG. Through a web interface developed with Streamlit, users can easily initialize, index, and query the knowledge base, creating a simple knowledge base Q&A system. This project aims to simplify the process of maintaining and querying a knowledge base, offering the following core features.
- Knowledge Base Management: Create, delete, and list existing knowledge bases.
- Configuration Editing: Edit the
.envandsettings.yamlconfiguration files through the web interface. - File Management: Upload and delete knowledge files (
.txtfiles). - Knowledge Base Initialization and Indexing: Initialize and index the knowledge base using
GraphRAGcommands. - Q&A Module: Query the knowledge base with support for various query methods (local, global, drift).
- 🐍 Python 3.11
- 🚀 Streamlit 1.27.0 or newer
- ⚙️ GraphRAG 0.4.0 or newer
- 📦 Dependencies specified in
requirements.txt
git clone https://github.com/CN-Scars/graphrag_web_ui
cd graphrag_web_uiUsing venv:
python3 -m venv graph_rag
source graph_rag/bin/activate # Unix/Linux
graph_rag\Scripts\activate # WindowsOr using Anaconda:
conda create -n graph_rag python=3.11
conda activate graph_ragpip install -r requirements.txtLaunch the Streamlit app:
streamlit run app-en.pyOpen your browser and visit http://localhost:8501 to use the GraphRAG Web UI.
- 🔧 Before using the indexing feature, ensure that GraphRAG is properly installed and configured.
- 📝 The
.envandsettings.yamlfiles can be manually modified or edited through the provided UI. - 📂 Knowledge bases are stored in the
knowledge_bases/directory. - 📁 Uploaded files are temporarily stored in the
uploads/directory.
Issues and pull requests are welcome to help improve this project.
If you have any questions, feel free to reach out!
GraphRAG Web UI
English | 简体中文