Chandra OCR 2 is a state-of-the-art optical character recognition model that converts images and PDFs into structured HTML, Markdown, or JSON while preserving layout information including complex tables, forms, handwriting, and mathematical equations.
This resource is most valuable when building applications that require digitizing physical documents, extracting structured data from PDFs, processing handwritten content, or converting complex layouts with tables and forms into machine-readable formats.