pdfpix

Turn a PDF into Markdown

or drop one here
Choose a PDF first.

Turn a PDF into checked Markdown without uploading it. Selectable text becomes lines, headings, lists and code, checked against the text kept from the PDF. PDFs with no selectable text export as Markdown with page pictures in a ZIP. Their lettering is not editable; OCR is needed to recover text.

How to turn a PDF into Markdown

  1. Choose a PDF

    Press Choose a PDF, or drop one on the page. The file is opened in this browser tab.

  2. Choose the structure

    Keep a marker for each PDF page or let the text flow continuously. Repeated headers and footers can be removed.

  3. Convert the text

    The tool reads the native text layer, rebuilds lines in reading order and infers simple headings, lists and code.

  4. Check and download

    The generated Markdown is read back and compared with the text kept from the PDF. A failed check does not produce a download.

What the conversion can preserve

A PDF records text at positions on pages. Markdown records a reading order and a small set of structures. This tool groups the native text into lines, keeps page boundaries when asked, and uses font size and font information to infer simple headings, lists and code. It does not claim to reproduce columns, complex tables, footnotes or visual placement exactly.

A PDF with no selectable text exports as page pictures, including scans and outlined letters. The ZIP holds document.md and a pages folder. Each Markdown image link points to a page in that folder. Unzip the download and keep them together. Every picture is decoded and checked before packaging, then the Markdown and image bytes are checked inside the ZIP. The lettering is not editable, and OCR is needed to recover text. A PDF with selectable text keeps the plain .md export.

Headings, headers and footers

Headings come from text size. The tool takes the median text size across the whole PDF. A line of 120 characters or fewer set at 1.35 times that size or larger becomes a level-two heading, and at 1.8 times or larger a level-one heading.

A line counts as a repeated header or footer when it sits in the top 18% or bottom 12% of the page, runs 160 characters or fewer, and repeats on at least 60% of pages, with a minimum of two. Letter case and extra spaces are ignored when lines are compared.

Questions

Does PDF to Markdown upload my file?
No. The PDF is opened, converted, checked and previewed in this browser tab.
Does this work on a scanned PDF?
Yes. If the PDF has no selectable text, it exports a ZIP containing document.md and a pages folder of 150 DPI page pictures. Unzip it and keep the file beside the folder. The lettering is not editable; use OCR to recover editable text. Outlined lettering uses the same visual export.
Will the Markdown look exactly like the PDF?
No. PDF stores positioned page content, while Markdown stores a reading order and simple structure. The tool preserves text and infers headings, lists and code where the PDF gives enough evidence.
What do page markers look like?
Each page starts with a Markdown comment such as Page 3. The comment keeps the page boundary without appearing as visible content in most Markdown viewers.
What happens to repeated headers and footers?
When the same short line appears near the same page edge on most pages, the tool can leave it out. The receipt says how many lines were removed.
Can the preview run links or HTML?
No. The Markdown preview is shown as plain text. It does not render HTML, follow links or load images.
What does the text check prove?
The tool reads the Markdown bytes back and compares their visible non-space character count with the PDF text it kept. At least 98% must remain, and replacement characters are refused.

Related tools

Each runs in a browser tab like this one, reads its own output back before the download, and needs no account. The footer links every tool.

Where to go next