GitHub-style Markdown, made in your browser
PDF to Markdown — headings, lists, tables and links as plain text
Turn a PDF into a .md file for GitHub, Notion, Obsidian, a docs site or an AI chat. Headings, bulleted and numbered lists, tables, code and links are marked up for you, two-column pages read in order, and repeated headers and page numbers are left out. Check the preview, then download or copy.
- PDF not uploaded to our server
- Tables as Markdown tables
- Ready for AI tools and docs
- Two-column pages in order
Convert PDF to Markdown
Add a PDF, check the preview, then download the .md file or copy the Markdown.
Drop a PDF here
or browse your files. Password-protected PDFs work too.
Your Markdown
Markdown settings
Leave empty for all, or type 1-3, 8-
Your PDF pages
Pages left out by the Pages box are dimmed. Click a page to compare it with the Markdown.
Extract, edit and reuse your content
Common next steps once your content is out of the PDF.
What this tool does
The words of your PDF, as plain Markdown.
Markdown is plain text with a few marks for structure: # for a heading, - for a list item, | for a table. It opens in any editor, and GitHub, Notion, Obsidian, documentation sites and AI chat tools all read it.
What you get
- Headings marked with # to ####, worked out from the text size.
- Bulleted and numbered lists, with nested items kept under their parent.
- Tables with clear columns as Markdown tables, numbers aligned right.
- Text set in a typewriter font, such as code, kept in a code block with its spacing.
- Links as [text](address), and a contents list from the PDF's bookmarks that links to the headings.
- Two-column pages read one column at a time, repeated headers and page numbers left out, and one UTF-8 .md file, so Gujarati, Hindi and every other script stay correct.
What it can't do
- Keep bold or italic inside a paragraph. The PDF's text doesn't say which words were bold.
- Bring the PDF's pictures, charts or colours. The .md file holds the text only.
- Rebuild tables without clear gaps between the columns, or cells merged across columns. Those come out as lines of text.
- Read scanned pages. They have no text; use PDF OCR first.
How to convert a PDF to Markdown in three steps
Open, check, save.
Open the PDF
Drop it on the box or press Select a PDF. If it is locked, type its password. The Markdown appears straight away.
Check the settings
- Which pages, and the reading order.
- Tables, links and the contents list.
- Page markers and front matter.
Check and download
Look at the Preview, or the Markdown tab for the text itself, then press Download .md, Copy Markdown, or Text (.txt) for the words only.
What people convert and what to set
Common jobs and the settings that suit each.
| Your PDF | Where the Markdown goes | What to set |
|---|---|---|
| A report or manual | ChatGPT, Claude or another AI tool, or a RAG knowledge base | Default settings. Headings and tables help the AI follow the document; press Copy Markdown and paste. |
| A guide or spec | A README or wiki page on GitHub or GitLab | Turn off Mark where each page starts, so paragraphs flow across pages. |
| Notes, articles or book chapters | Obsidian, Notion, Logseq or Bear | Turn on front matter to keep the title, author and dates as note properties. |
| A long manual with bookmarks | A docs site such as MkDocs or Docusaurus | Contents from bookmarks on; type 12-30 in Pages to take one chapter at a time. |
| A research paper in two columns | Your reading notes | Reading order Columns first. |
| A price list or schedule | A table you can edit as text | Tables as Markdown tables on; check the table in Preview. |
The details
How the Markdown is put together
A PDF doesn't say which line is a heading, a list or a table. This is how the tool works it out.
Headings
The most common text size is the body text. Short lines at least 15% larger become headings: the largest size gets #, the next ##, down to ####.
Lists
Lines starting with a bullet or dash become - items, and lines starting with 1. or 2) a numbered list. Short lines set in from the text around them are list items too, for PDFs that draw bullets as shapes.
Tables
Two or more lines in a row with wide gaps between their parts become a table. The first row is the header, and a column of numbers is aligned right with ---:.
Code and special characters
Lines in a typewriter font become a fenced code block. Characters that Markdown would read as formatting, such as * _ [ and |, get a backslash, so the text shows as it does in the PDF.
Links and contents
Web links in the PDF go on the words they cover, and addresses typed in the text are wrapped in angle brackets so they stay clickable. Bookmarks become a contents list whose entries link to the matching headings, the way GitHub names them.
Pages and columns
Each page starts with a hidden comment line that names it, such as Page 3; Markdown apps don't show it. Two columns are found by the empty strip between them, and the left one is read first.
When the Markdown isn't what you expected
What usually causes it, and what to change.
The Markdown is empty
The PDF pages are pictures, usually a scan, with no text.
Run it through PDF OCR first to get the words.
Columns mixed together
Reading order is Straight across, or the columns have no clear gap.
Choose Columns first. A page card says "columns" when they were found.
A heading came out as text
It is the same size as the body text, made to stand out only by bold or colour.
Add ## and a space at the start of that line in the .md file.
A table came out as lines
Its columns sit close together, or it has only one row, or Tables as Markdown tables is off.
Turn the setting on. For a tight table, PDF to Excel may read it better.
Odd letters or boxes
The PDF's fonts don't say which letters their shapes are, common in older Indian-language PDFs.
Use PDF OCR, which reads the letters from the page pictures.
Better Markdown on the first try
Small choices that save editing later.
Look at the Preview first
It shows the headings, lists and tables the tool found, drawn the way Markdown apps show them.
Turn page markers off for articles
Paragraphs then flow across pages, and the file has no page comments.
Send AI tools only what matters
Type the pages you need, such as 3-8, so the chat isn't filled with covers and appendices.
Keep headers and footers out
Leave that setting on so a title repeated on every page doesn't break up the text.
Use front matter for notes apps
Obsidian and static site tools read the title and dates from it.
Keep the original PDF
Converting again with other settings is quicker than fixing Markdown by hand.
Common questions about PDF to Markdown
If something's not covered here, the sections above likely have it.
What does PDF to Markdown give me?
One .md file with the PDF's text marked up as Markdown: # for headings, - and 1. for lists, | for tables, fenced code blocks for text in a typewriter font, and [text](address) for links. You can download it, copy it, or save a plain .txt file with the words only.
Will the Markdown look exactly like my PDF?
No. Markdown keeps the structure, not the design. Fonts, colours, pictures and the position of things on the page are left out, and bold or italic inside a paragraph isn't kept because the PDF's text doesn't say which words were bold. For the exact look, use PDF to HTML instead.
Does it convert tables?
Yes, when the columns have clear gaps between them. Two or more such rows become a Markdown table with the first row as the header, and columns of numbers are aligned right. Tables with cells merged across columns or tightly packed columns come out as lines of text; turn off Tables as Markdown tables if you prefer that for every table.
Can I use the Markdown with ChatGPT, Claude or other AI tools?
Yes. Markdown keeps headings, lists and tables in a form AI tools read well, and it is smaller than the PDF. Press Copy Markdown and paste it into the chat, or type only the pages you need in the Pages box first.
Is my PDF uploaded anywhere?
No. The PDF is read with pdf.js in this browser tab and the Markdown is made here too; the file isn't uploaded to our server. Closing the tab clears it.
What are the Page comments in the file?
Each page of the PDF starts with a hidden comment line that names it, such as Page 3. Markdown apps don't display them, but they help you find a passage in the original. Turn off Mark where each page starts to leave them out, which also joins paragraphs that run across pages.
Why is the Markdown empty or missing some pages?
Those pages are pictures, usually a scan, so there is no text to read. The file notes each such page in a comment. Run the PDF through PDF OCR first, then convert the result.