Interactive Demo

Retrieval-Augmented Generation

This project demonstrates how RAG works end-to-end: retrieving relevant content from a text corpus using different search strategies, then using that context to power AI-generated responses.

Pick one of the two modules below to get started. Each one focuses on a different part of the RAG pipeline.

Search & Retrieval

See how different retrieval strategies find relevant content from a body of text.

How to use

  • 1Paste any text (or load the included Harry Potter sample data)
  • 2Type a search query and press Enter
  • 3Switch between Vector, BM25, and Combined (RRF) retrieval methods
  • 4Compare how each method ranks and surfaces different snippets
  • 5Adjust the result count (5, 10, or 20) to control output volume
Vector SearchBM25Hybrid RRF

Chat Interface

Ask questions in natural language and get AI-generated answers grounded in retrieved context.

How to use

  • 1Start a new chat session
  • 2Type a question or prompt and press Enter
  • 3View the AI response rendered in a clean conversational layout
  • 4Continue the conversation with follow-up questions
  • 5Start a fresh session anytime with the New Chat button
ConversationalRAGReal-time

This is a demo project for educational purposes only. Requests may be throttled. View source on GitHub