metadata

title: Easy RAG
emoji: 🐢
colorFrom: purple
colorTo: indigo
sdk: gradio
sdk_version: 5.8.0
app_file: app.py
pinned: false
license: mit
short_description: Chat with your docs very Easy

Easy RAG 🤖

Easy RAG is a powerful and user-friendly Retrieval Augmented Generation (RAG) system that allows users to upload their own documents and query them using state-of-the-art language models.

Features

📄 Support for multiple document formats (PDF, TXT, DOCX)
📚 Upload up to 5 documents (max 10MB each)
🔍 Advanced document processing and chunking
💡 Intelligent question answering using Llama-2
🌐 Multilingual support
🚀 GPU-accelerated inference
📊 Source tracking and citation

Technical Stack

Language Model: Meta-llama/Llama-2-7b-chat-hf
Embeddings: intfloat/multilingual-e5-large
Vector Store: FAISS
UI Framework: Gradio
Document Processing: LangChain

Installation

Clone the repository
Install dependencies:

pip install -r requirements.txt

Set up your HuggingFace token as an environment variable:

export HUGGINGFACE_TOKEN=your_token_here

Run the application:

python app.py

Usage

Upload your documents using the file upload interface
Wait for the system to process and index your documents
Start asking questions about your documents
View answers with source citations

Requirements

See requirements.txt for a complete list of dependencies.

Credits

Based on original work by Camilo Vega, AI Professor and Solutions Consultant.

License

MIT License