MarkItDown is a lightweight Python tool for converting various files to Markdown format for use in LLM and related text analysis processes. It is most similar to Text, but focuses more on retaining important structure and content of the document (including titles, lists, tables, links, etc.). Although the converted document is usually beautiful and easy to read, it is mainly aimed at text analysis tools and may not be the best choice for users who need high-fidelity document conversion for human reading. MarkItDown currently supports conversions to the following formats: PDF PowerPoint Word Excel Image (EXIF metadata and OCR) Audio (EXIF metadata and voice transcription) HTML Text formats (CSV, JSON, XML) ZIP file (traverse content) EPub ... and more!
