ocr-and-documents

Use pymupdf to open the PDF and call get_text() on each page, or run the extract_pymupdf.py script for plain text, markdown, tables, or metadata. It installs in seconds at about 25MB with no model downloads.

Use pymupdf for text-based PDFs, basic tables, and image extraction since it is fast and lightweight. Use marker-pdf when you need OCR for scanned documents, equation and LaTeX recognition, forms, or reading-order detection, accepting its 3-5GB install size.

No, pymupdf cannot perform OCR on scanned PDFs because it only reads existing text layers. For scanned documents, use marker-pdf, which supports OCR across 90+ languages, or try web_extract if the document has a public URL.

Use pymupdf's insert_pdf method to copy selected pages into a new document for splitting, or combine multiple source documents into one for merging. No extra dependencies beyond pymupdf are required.

marker-pdf requires roughly 5GB of disk space for PyTorch plus about 2.5GB of models downloaded to the HuggingFace cache on first use. Run the script's --check flag to verify free disk space before installing.