--- title: RAG Chat Flow emoji: 🚀 colorFrom: gray colorTo: gray sdk: streamlit sdk_version: 1.29.0 app_file: app.py pinned: false license: mit short_description: 'RAG-powered document intelligence with vector search and AI ' --- # RAG Chat Flow 📚 An intelligent document Q&A chatbot powered by Retrieval-Augmented Generation (RAG). Upload your documents and ask questions to get accurate, context-aware answers. ## Features - 🔍 **Smart Document Search**: Uses semantic search to find relevant information - 🤖 **AI-Enhanced Answers**: Combines document retrieval with AI processing for natural responses - 📁 **Multi-Document Support**: Upload multiple text files for comprehensive knowledge base - 💬 **Chat History**: Persistent chat sessions with automatic saving - 👥 **Multi-User Support**: Real-time user tracking and session management - 🎯 **Confidence Scoring**: Shows how confident the system is in its answers - 📊 **Source Attribution**: Always shows which documents answers come from ## How to Use 1. **Upload Documents**: Use the sidebar to upload .txt files containing your knowledge base 2. **Index Documents**: Click "Re-index Documents" to process your files 3. **Ask Questions**: Start chatting! Ask specific questions about your uploaded documents 4. **Get Answers**: Receive both AI-enhanced and extracted answers with source citations ## Technical Details - **Embedding Model**: all-mpnet-base-v2 for semantic understanding - **Vector Database**: ChromaDB for efficient similarity search - **Text Splitting**: Intelligent chunking that preserves context - **AI Enhancement**: Optional OpenRouter API integration for natural language responses ## Environment Variables Set `OPENROUTER_API_KEY` for AI-enhanced responses (optional - works without it too). ## Example Use Cases - Corporate policy documents - Technical documentation - Research papers - Product manuals - Legal documents - Knowledge bases ## Setup The app automatically handles document processing and indexing. Simply upload your text files and start asking questions!