README.md

metadata

title: RAG
emoji: 🔍
colorFrom: yellow
colorTo: green
sdk: gradio
sdk_version: 5.30.0
app_file: app.py
pinned: false
license: mit
short_description: Retrieval-Augmented Generation (RAG)
models:
  - bert-base-uncased
  - google/flan-t5-base

📄🔍 Retrieval-Augmented Generation (RAG) Demo

A simple yet powerful RAG application that lets you upload documents and ask questions about them.

Features

• 📄 Upload multiple .txt files
• 🔍 Automatic document processing and indexing
• 💡 Query your documents using natural language
• 🤖 Get AI-generated answers based on your content

How It Works

1. Upload - Add your text files to the system
2. Index - Documents are embedded using bert-base-uncased
3. Query - Ask a question about the documents
4. Retrieve - The system finds the most relevant content using cosine similarity
5. Generate - flan-t5-base creates a natural language answer

Technical Details

• Built with Hugging Face's Transformers
• Uses cosine similarity for matching
• No GPU required (ZeroGPU compatible)
• Runs completely in-memory

Usage

Simply upload your text files, ask a question, and get an answer within seconds!

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference