How to convert a PDF to Markdown
This tool reads the text in your PDF and gives it back as a .md file: paragraphs as plain text, headings marked with # and ##, and tables written as Markdown tables. It is what you want when a manual, report or set of minutes needs to go into a Git repository, a wiki such as Obsidian or Notion, or an LLM prompt without all the noise of the PDF format.
It is not a visual conversion. Markdown has no positions or fonts, so the goal is clean, well-structured text rather than a copy of the layout.
Step by step
- Upload one or more PDFs; with several, each is converted separately and you get them together in a ZIP.
- If the PDF is a scan with no text, the tool offers to run OCR and convert in the same job.
- Wait for processing to finish and download the .md file.
- Open it in your editor and check headings and tables before publishing.
How the structure is rebuilt
A PDF never says 'this is a heading'; it only stores characters with a size and a position. Headings are therefore inferred by comparing each line with the usual font size on the page: lines that are clearly larger become #, slightly larger ones become ##, and everything else stays a paragraph.
Detectable tables, the ones with recognisable rows and columns, are written with Markdown pipe syntax. Two-column pages are read one column at a time instead of interleaving lines. Each page starts with an <!-- Página N --> comment so you can trace every fragment back to its page.
What is lost and what to check
Images, colours, fonts and exact positions do not carry over. Linked footnotes and merged table cells cannot be reproduced either, because Markdown has no syntax for them. Pay most attention to headings set in unusual sizes and to tables without borders.
The limit is 50 pages per PDF and 50 MB per job. For a longer document, split it first with Split PDF and convert the parts.
Frequently asked questions
- Does it work with scanned PDFs?
- Yes, but a scan is an image with no text. The tool detects this and offers to run OCR and convert in one step, without uploading the file again.
- Do tables come out as Markdown tables?
- Detectable ones do, with a header row and separators. Borderless tables or tables with merged cells may need touching up.
- Are the PDF's images included?
- No, the output is text only. To get the images, use PDF to JPG with the embedded images option.
- Why convert a PDF to Markdown?
- To edit the content as text, version it in Git, publish it on a wiki or pass it to an AI with far less noise than the original PDF.
- Can I convert several PDFs at once?
- Yes. Each PDF is converted separately with the same options and the download arrives as a ZIP.