🧠 RAG Chatbot with LLaMA 3 & Gradio

A lightweight and interactive chatbot powered by Meta’s LLaMA 3 and enhanced through Retrieval-Augmented Generation (RAG) for domain-specific answers. Comes with a clean Gradio UI for easy access and testing.

---

🚀 Features

• 🔍 RAG architecture for better factual accuracy
• 🤖 Powered by Meta’s LLaMA 3 model
• 🎛️ Easy-to-use Gradio interface
• 📚 Plug in your own data (PDFs, text files, etc.)
• 🌐 Local or cloud-hosted deployment

---

📸 Demo Screenshot

Case Study: We provide specific details in plain text of an object and ask questions regarding its properties. The LLM must reply only based on the provided context.

---

📁 Project Structure

rag-llama3-gradio/
├── main.py                # Entry point: runs the Gradio app
├── loader.py              # Loads and chunks documents for processing
├── embedder.py            # Handles embeddings and indexing using ChromaDB
├── rag_engine.py          # Retrieval + Generation logic using LLaMA 3
├── requirements.txt       # Python dependencies
├── README.md              # Project documentation
├── data/                  # Source documents for retrieval
│   └── demo.txt           # Sample file for testing
└── .gitignore             # Files and folders to exclude from Git

---

🛠️ Installation

git clone https://github.com/raf-init/rag-llama3-gradio.git
cd rag-llama3-gradio
pip install -r requirements.txt