PDF to Markdown
Turn a PDF into clean Markdown, with its headings, lists, tables and links, ready for an AI tool, GitHub or your notes.
or drop it here · up to 50 MB, free
Bigger files? Pro converts files up to 500 MBOver the free limits: you can unlock the download from €1.99.
Working on it…
Your file is ready
Free · No sign-up · No watermark · Your files never leave your device
PDF to Markdown in 3 steps
- Select your PDF, or drop it on the box.
- Choose whether to keep the images, then click Convert to Markdown.
- Download the .md file and open it in any editor, or paste it into ChatGPT, Claude or Obsidian.
Why Markdown is the best way to give a PDF to an AI
Copying text out of a PDF gives you broken lines, page numbers in the middle of sentences and tables that fall apart into loose words. This converter reads the layout of every page instead: bigger text becomes # headings at the right level, bold lines on their own become subheadings, bullets and numbered lists stay lists, tables become GitHub-flavoured Markdown tables and web links keep their address. Two-column pages are read in the right order, words split by a hyphen at the end of a line are joined again, and a paragraph cut by a page break comes out whole.
Running headers, footers and page numbers that repeat on every page are left out, so the result is just the document. Monospaced text such as code or commands is kept in code blocks with its indentation. Scanned pages without a text layer are read with OCR in the language of this page and in English. Markdown has no fonts, colours or page layout, so those are left out on purpose; for an editable copy that looks like the PDF, use PDF to Word.
Frequently asked questions
How do I convert a PDF to Markdown?
Select the PDF, choose whether you want the images and click Convert to Markdown. You get a .md file, or a ZIP with the .md file and a folder of images if you chose to keep them. Open it in VS Code, Obsidian, Typora or any text editor.
Is Markdown better than plain text for ChatGPT or Claude?
Yes. Language models understand Markdown very well, and headings, lists and tables tell them how the document is organised. With plain text copied from a PDF, tables lose their columns and the model has to guess where sections start, so the answers are usually less precise.
Are tables converted?
Yes. Tables with ruled lines and tables made of aligned columns become Markdown tables with the same rows and columns, which GitHub, GitLab, Obsidian and most editors show as a grid. Markdown tables cannot merge cells, so the text of a merged cell goes in its first column.
Does it work with scanned PDFs?
Yes. Pages without selectable text are read with OCR inside your browser, in the language of this page and in English, and come out as normal Markdown paragraphs and headings. Clear, straight scans give the best text, so check names and numbers afterwards.
What happens to images and charts?
By default they are left out, so you get a single .md file that is easy to paste or upload. Choose In a folder to save the pictures as image files next to the Markdown, linked from the text, and download both in a ZIP. Logos repeated on every page are only kept once.
Is my PDF uploaded to a server?
No. The PDF is read and converted inside your browser on your own device, so contracts, reports and research papers are never sent to us or stored anywhere. Once the page has loaded it keeps working even offline.
Related tools
Part of: PDF tools