paddleocrPaddlePaddle
ADDEDPaddleOCR is an open-source OCR toolkit that converts PDFs and images into structured data (JSON/Markdown) for LLMs. It includes PP-OCRv6 for multilingual text recognition (100+ languages), PaddleOCR-VL for document parsing, and PP-StructureV3 for layout-aware conversion. Supports CPU, GPU, and various AI accelerators.