How to convert a PDF to Markdown in 3 steps
- Open the PDF. Select it or drop it here, and type its password if it has one.
- Pick pages and options. Choose whether to include pictures, leave out page headers and footers, or mark where pages start.
- Convert, then copy or download. The Markdown appears in a preview with a Copy button, and as a .md file to download.
What you get
Pasting text out of a PDF gives you a line break at the end of every line and loses every heading. Here the structure comes back:
- Headings from the type sizes: the biggest titles become
#, the next##, and so on. - Paragraphs with their lines joined again and words broken by a hyphen mended.
- Bold, italic and links as
**bold**,*italic*and[text](address). - Lists, bulleted and numbered, nested by their indent.
- Tables as Markdown tables (the kind GitHub, Obsidian and most editors show).
- Code set in a typewriter font as fenced code blocks, with its indents.
A short example. A page with a title, a sentence and a two-column table comes out like this:
# Quarterly results
Sales grew in **every region** this quarter.
| Region | Revenue |
| --- | --- |
| North | $24,000.00 |
| South | $19,000.00 |
The Markdown you’ll see, at a glance
| In the PDF | In the Markdown |
|---|---|
| Largest title | # Title |
| Section and sub-section headings | ## Section, ### Sub-section |
| Bold and italic words | **bold**, *italic* |
| A link | [text](https://address) |
| Bullets and numbered steps | - item, 1. step, indented when nested |
| Text in a typewriter font | a block between ``` lines |
| A table | rows of cells between | signs |
| A line break kept on purpose (an address) | two spaces at the end of the line |
Reading order that makes sense
A PDF stores text in whatever order it was drawn, which is often not the order you read it. The converter works out the layout first: two- and three-column pages are read one column after the other, a title over both columns comes first, and text that runs from the foot of one column to the top of the next becomes one paragraph again.
Tables
Tables drawn with lines and tables whose columns are only lined up with spacing both become Markdown tables. Markdown tables have exactly one header row and no merged cells, so a cell spanning several columns keeps its text in the first of them; check very wide or complex tables. For tables you want to calculate with, PDF to Excel gives real numbers.
Images: leave them out, or get a folder
For notes and AI chats you usually want text only, so pictures are left out by default and you get one .md file. Choose Include pictures to download a ZIP instead: the Markdown file plus an images folder, with each picture linked where it appeared, such as .
Page headers, footers and page numbers
Running headers, footers and page numbers repeat on every page and break up paragraphs. They are left out by default. If you need to know where each page started, tick Mark where each page starts to add a comment such as <!-- Page 3 -->, which is invisible when the Markdown is shown. To convert only part of a long document, select those pages before converting.
Markdown for ChatGPT, Claude and other AI tools
AI chats read Markdown well: headings show what each part is about, lists stay lists and tables keep their rows. A clean Markdown copy of a report, contract or paper is easier for the assistant to follow than a messy paste with broken lines. Choose the pages you need to keep it within the chat’s limits, then press Copy Markdown.
Markdown for Obsidian, Notion, GitHub and docs sites
Drop the .md file into your Obsidian vault, a GitHub repository or a docs folder, or import it into Notion. With pictures included, unzip the download and keep the images folder next to the .md file so the links work.
Private by design
The PDF never leaves your device: it is read and converted in your browser. That matters for contracts, medical letters and internal documents you might otherwise upload to a converter and then an AI tool. Password-protected PDFs open with your password, on your device; to save a copy without the password, use Unlock PDF.
Scanned PDFs
Try to select a line of text in your PDF viewer. If the whole page highlights like a photo, it is a scan: a picture of text. Reading a picture needs text recognition (OCR), which this tool doesn’t do. Rather than giving you an empty file, it tells you which pages are scans.
No file size, page or daily limits
Many online converters stop at 15 MB, 30 pages or a few files, or charge for long documents. Here there is no counter and no sign-up. The only limit is your device’s memory; on a phone, split a very long PDF first.
Other ways
- Copy and paste from a PDF viewer keeps the words but loses headings, lists and tables, and adds a line break on every line.
- Command-line converters work well for developers but need installing and setting up.
- Pandoc converts between many formats but can’t read PDF files.
- Uploading the PDF to an AI chat works for some files, but the PDF goes to their servers, and long or complex files may be cut short. For plain text without any Markdown, use PDF to Text. For an editable document, use PDF to Word.
Frequently asked questions
Is my PDF uploaded?
Is there a limit on file size, pages or conversions?
Will headings, tables and links be kept?
# headings by level, tables become Markdown tables, links become [text](address), and bold, italic, lists and code blocks are kept. Check tables with headings stacked over several rows, since Markdown tables have one header row.Can it handle two-column PDFs like research papers?
What happens to images?
images folder it links to.Can I convert a scanned PDF?
Can I convert a password-protected PDF?
Can I use the Markdown in ChatGPT, Claude, Obsidian or Notion?
Last updated: 2026-11-20