This project focuses on building a system for automatically generating assessments from scanned book pages, using advanced OCR (Optical Character Recognition) and AI models. The system processes uploaded images or PDFs, extracts text via PaddleOCR, and generates tailored educational content using LLaMA 3.2. The system uses the RAG (Retrieval-Augmented Generation) principle and integrates LangChain for seamless orchestration of data and AI models.