MarkItDown: turn documents into Markdown for LLM pipelines
MarkItDown is a Microsoft open-source Python tool that converts PDFs, Office files and more into clean Markdown for LLM workflows.
What it is: MarkItDown is an open-source Python utility from Microsoft. It converts files such as PDFs, Word documents, Excel sheets and PowerPoint decks into Markdown.
Who it is for: developers building AI features that need to read documents, for example search over company files or a chat with a PDF.
Why it is worth starring: language models handle Markdown well, and it keeps useful structure such as headings, lists and tables. Converting messy documents into Markdown first usually makes the next steps, like splitting text into chunks, simpler.
pip install 'markitdown[all]'from markitdown import MarkItDown
md = MarkItDown()
result = md.convert("report.pdf")
print(result.text_content)One limitation: the output is designed for machines to read, not for a perfect human-readable copy of the original. If you need exact layout and formatting, it is not the right tool.