🤖 RAG-Based LLM Chatbot

A Retrieval Augmented Generation based AI chatbot built using a fully open source stack.
This application allows users to upload PDF documents, generate embeddings, and interact with their documents through an intelligent chatbot all running locally for privacy and control.

---

📌 Project Overview

The RAG-Based LLM Chatbot is a Streamlit-powered web application that enables document-based question answering.
By combining semantic search with a local Large Language Model (LLM), the chatbot provides accurate, context-aware answers grounded in uploaded documents.

---

✨ Features

• 📂 Upload PDF Documents
  Upload and preview PDF files directly in the app.

• 🧠 Create Embeddings
  Generate semantic embeddings for documents using BGE embeddings.

• 🗄️ Vector Storage with Qdrant
  Store and retrieve document embeddings efficiently using a local Qdrant vector database.

• 🤖 Intelligent Chatbot
  Ask natural language questions and get accurate answers from your documents.

• 🔐 Local & Private
  No external APIs required — runs completely on your machine.

• 🖥️ User-Friendly UI
  Clean and intuitive Streamlit interface.

---

🧠 RAG Architecture (How It Works)

1. PDF Upload → User uploads a document
2. Text Extraction + OCR → Text extracted using unstructured
3. Embeddings Creation → BGE embeddings generated
4. Vector Storage → Stored in Qdrant
5. Query Embedding → User question embedded
6. Semantic Retrieval → Relevant chunks retrieved
7. Answer Generation → LLaMA 3.2 generates a grounded response

---

🛠️ Tech Stack

Core Technologies

• Python 3.10
• Streamlit – UI framework
• LangChain – RAG orchestration
• Qdrant – Vector database (Docker-based)
• Ollama – Local LLM runtime

Models & Libraries

• LLaMA 3.2 (via Ollama) – Local LLM
• BGE Embeddings (BAAI/bge-small-en) – Semantic embeddings
• sentence-transformers
• unstructured + OCR (Tesseract) – PDF text extraction

---

---

🚀 Getting Started

Clone the Repository

---

# ================================
# RAG-Based LLM Chatbot Setup
# ================================

# 1. Clone the repository
git clone https://github.com/padmavathi2905/LLM-Chatbot.git
cd LLM-Chatbot

# 2. Create and activate virtual environment (Python 3.10 recommended)
python3.10 -m venv venv --upgrade-deps
source venv/bin/activate

# 3. Upgrade pip tools
python -m pip install --upgrade pip setuptools wheel

# 4. Install Python dependencies
python -m pip install -r requirements.txt

# 5. Install Unstructured with full PDF + OCR support
python -m pip install "unstructured[pdf,image,ocr]"

# 6. Fix NumPy compatibility (LangChain requires numpy < 2)
python -m pip install "numpy<2"

# 7. Install OpenCV (NumPy-compatible, headless)
python -m pip install "opencv-python-headless<4.9"

# 8. Install PDF image conversion dependency
python -m pip install pdf2image

# 9. Install system dependencies (macOS)
brew install poppler tesseract

# 10. Set OCR engine
export OCR_AGENT=tesseract

# 11. Pull LLaMA model using Ollama
ollama pull llama3.2:3b

# 12. Start Qdrant using Docker
docker run -d \
  --name qdrant \
  -p 6333:6333 \
  qdrant/qdrant

# 13. Run the Streamlit application
python -m streamlit run new.py