Convert a PDF to Markdown
Pull the text out as clean Markdown — headings, lists, tables and links kept — ready for Obsidian, Notion or a repo.
Free · No signup · No watermark · Your file never leaves your device.
Drop your PDF here
or click to browse — the text is read on your device, nothing is uploaded
How to PDF to markdown in 4 steps
- 1
Open your PDF
Drop the file onto the box. The text layer is read on your own device — the document is never uploaded.
- 2
Choose what to look for
Headings, tables, links and the removal of repeated headers are all on by default. Turn any of them off if your document confuses them, and pick a page range if you only want part of it.
- 3
Convert
Every page is read in turn and the layout is translated into Markdown: bigger lines become headings, bulleted lines become lists, aligned columns become tables.
- 4
Copy or download
Check the result in the preview tab, then copy the Markdown straight into your notes or save it as a .md file.
What a PDF actually contains
It is worth knowing what this conversion is doing, because it explains both what it is good at and where it can be wrong. A PDF does not store a document in the way a word processor does. There are no paragraphs in the file, no headings, no lists — only instructions to draw particular glyphs at particular positions on a page of a particular size.
The structure you see when you read a PDF is something you supply. A line set in larger type at the top of a section reads as a heading because that is what large type in that position has always meant. A short line starting with a dot reads as a list item. A converter has to work from the same evidence you do: size, position, spacing, alignment and repetition.
What comes across well
Ordinary documents — reports, papers, manuals, ebooks, minutes — convert cleanly. Paragraphs are rejoined from the lines they were broken into, including across a page break, and words split by a hyphen at the end of a line are put back together. Headings, bulleted and numbered lists, simple tables and clickable links all survive. Page numbers and the header repeated on all forty pages are dropped, because they are furniture belonging to the paper, not to the text.
What does not
Inline emphasis is the main casualty: the text layer says which font a run used, not that the author meant it to stand out. Complex tables with merged cells will come across as approximations. Figures and equations set as images are not text and will simply be absent. And a scanned document contains no text at all — you will be told so rather than handed an empty file.
Why convert at all
Markdown is text. It can be searched, diffed, version controlled, pasted into a note-taking app, rendered into a web page, and read by anything that reads files. A PDF is a fixed picture of a page, which is exactly what you want when you are printing it and exactly what you do not want when you are trying to quote three sentences from it into your notes.
PDF to Markdown — frequently asked questions
Are my files uploaded to a server?+
No. The PDF is opened and read inside your browser, and the Markdown is built on your device. Nothing is sent anywhere, which also means the tool keeps working if you disconnect from the internet after the page has loaded.
Why does it not work on my scanned document?+
A scan is a photograph of a page. There are no words in the file, only pixels arranged to look like words, so there is no text to convert. Recovering text from a picture needs OCR, which this tool does not do — you will see a message saying no text was found rather than an empty file.
How does it know what is a heading?+
A PDF does not record headings. It records glyphs at coordinates, and you recognise a heading because it is bigger and sits in its own whitespace. The converter reads the same evidence: text set noticeably larger than the body becomes a heading, with the largest size in the document becoming #, the next ## and so on.
Will bold and italic survive?+
Usually not, and this is the honest limitation of any converter that reads the text layer. The layer records which font each run was set in, not what the author meant by it, so there is no dependable way to tell a bold run from one that merely used a different typeface. Structure — headings, lists, tables, links — is recovered; inline emphasis generally is not.
How good is the table detection?+
It is deliberately cautious. Columns are only rebuilt as a Markdown table when several consecutive lines share the same column positions and the cells are short. That means a genuine table usually comes across, and a page of prose is never mangled into one. If a table in your document is missed, its rows will still be there as text.
What happens to two-column documents?+
They are read one column at a time, not straight across. The page is checked for a channel of whitespace running down it, and if the text either side of that channel behaves like a column — filling its width, running most of the height of the page — the two are read in order rather than interleaved line by line.
Can I convert only part of a long PDF?+
Yes. Type a page range like 1-5, 9 in the Pages box and only those pages are read. This is worth doing on a very long document: it is faster, and it keeps the output to the part you actually want in your notes.
What can I do with the Markdown afterwards?+
Paste it into Obsidian, Notion, Bear or a GitHub repository; feed it to a static site generator; or keep it as a plain text copy that will still open in fifty years. Markdown is also far cheaper than a PDF to hand to a language model, because it is text and structure with nothing else attached.
Other free PDF tools
JPG to PDF
Turn photos and scans into one PDF. Set page size, orientation and margin.
PDF to JPG
Turn every page into a picture you can post, print or drop into a slide.
Merge PDF
Combine two or more PDFs into a single document. Drag to reorder, then download — all in your browser.
Split PDF
Break one PDF into several documents — by page range, or one file per page.