Launch offer20% off every Pro version until 30 November, applied at checkoutSee the pricesHoliday offer15% off every Pro version until 31 December, applied at checkoutSee the prices
mutewareTeam plans

Free tools / PDF to Markdown

PDF to Markdown

Choose a PDF and get clean Markdown: headings from the font sizes, paragraphs, lists and simple tables, with running headers and page numbers left out. The file is read on your own computer and never uploaded, so it is fine for contracts, reports and other documents you would rather not send to a converter site.

Is there personal data in what you are about to paste into an AI chat? Scrub it first, also in your browser.

How it works

Open the PDF

Choose a file or drop it on the box. pdf.js, the PDF reader of Firefox, reads the text on this page.

Check the Markdown

Headings, lists and tables are recognised from the layout. Edit anything in the box before you copy.

Copy or download

Paste it into ChatGPT, Claude, Obsidian or GitHub, or save it as a .md file.

Questions

Is my PDF uploaded?

No. The file is opened by JavaScript on this page and read on your own computer. Nothing is sent to us or anyone else, and there is no account. Once the page has loaded it keeps working without an internet connection.

Why convert a PDF to Markdown before giving it to an AI model?

Markdown keeps the structure (headings, lists, tables) as plain text, so a model can follow the document and you can see exactly what it gets. You can also cut the parts it does not need, which saves tokens, and check that no personal data goes along.

How are headings recognised?

From the font size. The size used for most of the text is taken as body text; lines in clearly larger sizes become headings, the largest size level 1, the next level 2 and level 3. Short lines in bold become bold text.

Does it handle tables?

Simple ones. When several lines in a row have text in the same number of columns with clear space between them, they become a Markdown table. Tables with merged cells or text that wraps inside a cell are not always recognised; check them and fix them in the box, or use the table converter on the cells.

Does it work on scanned PDFs?

No. A scan is a picture of a page without a text layer, and this tool reads text, not images. It tells you when that is the case. Run the scan through OCR first, for example in your scanner software, and open the result here.

What is left out?

Images, colours and page layout. Running headers, footers and page numbers that come back on most pages are removed too; tick the option to keep them.

More free tools

Do this on any page in one click

Copy for AI turns a selection or a whole web page into clean Markdown, without menus, ads or tracking links. Free, no account.