🚀 Documentation crafted by Sonicar Tech LLC — get your own docs shipped in 1 week →

Markdown for LLM pipelines

Turn any document into clean, LLM-ready Markdown

MarkItDown is a lightweight Python utility that converts PDF, Office documents, images, audio, HTML and more into Markdown — preserving the document structure that text-analysis tools rely on.

$ pip install 'markitdown[all]'

Every format you have

PDF, Word, Excel, PowerPoint, images, audio, HTML, CSV, JSON, XML, ZIP, EPub and YouTube URLs.

Structure preserved

Headings, lists, tables and links survive the conversion — and Markdown is token-efficient.

CLI, Python or Docker

Pipe a file through the CLI, call the Python API, or run the container with no local install.

Plugin-extensible

Add formats and LLM-powered OCR with 3rd-party plugins, without forking the core repo.

Azure AI integrations

Route conversions through Document Intelligence or Content Understanding for higher fidelity.

Security-aware by design

Narrow convert_* entry points let you give MarkItDown exactly the reach it needs.