Skip to content

Latest commit

 

History

History
52 lines (38 loc) · 1.62 KB

README.md

File metadata and controls

52 lines (38 loc) · 1.62 KB
title emoji colorFrom colorTo sdk sdk_version app_file pinned
RAGio
📉
indigo
indigo
gradio
4.27.0
app.py
false

RAGio - simple to start RAG with HuggingFace gradio interface.

Retrieval-Augmented Generation (RAG) with a Gradio interface. Perfect for both beginners and experienced developers looking to integrate advanced NLP features.

Features

  • Beginner Friendly: Easily set up and run your RAG models locally or host them on HuggingFace Spaces. Currently supports *.pdf documents only.
  • Interactive UI: Engage with your models and data in real time through a dynamic Gradio interface.
  • HuggingFace and OpenAI APIs: Utilizes HuggingFace and OpenAI API.
  • Vector store with LanceDB: Utilizes LanceDB to store and manage embedding vectors.
  • Multiple Chunking Strategies: Coming soon ...

Quick Start

Get started with RAGio by cloning the repository and installing dependencies:

git clone https://github.com/your-username/RAGio.git
cd RAGio
pip install -r requirements.txt

Run the application

Important!
Ensure to configure your environment by filling in the .template.env file with your HuggingFace and OpenAI credentials. Rename this file to .env after updating.

cd RAGio
source ./.env  # apply environment variables
gradio app.py  # run gradio app

Open http://127.0.0.1:7860 in your browser.

For more details on configuration and usage, check out our documentation.

Contributing

Contributing Your contributions are welcome! If you have suggestions or want to improve RAGio, feel free to fork the repository, make changes, and submit a pull request.