Skip to content
ToolMug

Use ↑ ↓ to move, Enter to open, Esc to close.

PDF to Markdown

Dropping a raw PDF into an AI chat often mixes in headers, footers, page numbers and broken lines. Convert it to Markdown first — with headings, lists and paragraphs cleaned up — and the AI answers more accurately while using fewer tokens.

Runs in your browser — nothing you enter is uploaded

or drop a file here

Headings are inferred from font size, lists from bullets and numbering. Tables and multi-column layouts come out as plain text, so check them before sending.

How to use

  1. Choose a PDF and wait while each page is read.
  2. Tick Remove headers, footers and page numbers or Separate pages with --- as needed.
  3. Copy the result into your AI chat, or download the .md file.

Options explained

Headings
Lines at least 15% larger than body text become headings, in up to three levels by size.
Headers and footers
Text that repeats in the top or bottom two lines on more than half of the pages is removed, as are lines holding only a page number.

What changes on the way to Markdown

The converter reads each piece of text with its position and font size, then rebuilds the structure. Take a three-page report where every page carries a running header and a page number:

Company Name                      9 pt, top of every page
Annual Report                     24 pt, page 1 only
Revenue grew steadily during the year, driven by
new customers in three regions.   12 pt body text
- First item
- Second item
Page 1                            9 pt, bottom of every page

With the default settings, page 1 comes out as:

# Annual Report

Revenue grew steadily during the year, driven by new customers in three regions.

- First item
- Second item

The two body lines are joined into one paragraph, the 24 pt line becomes a heading because it is more than 15% larger than the 12 pt body text, and the header and page number are dropped. Header and footer removal needs a document of at least three pages and only looks in the top and bottom 10% of each page. A gap between lines of nearly twice the font size starts a new paragraph.

Frequently asked questions

Does it work on scanned PDFs?

No. Text in a scan is an image and needs OCR; this tool only reads the PDF's text layer.

Are tables kept?

They are not turned into Markdown tables; their content becomes plain lines. Multi-column layouts may also come out in the wrong order, so check before sending.

Is my file uploaded?

No. Everything happens in your browser.

Privacy

This tool runs entirely in your browser. Your files and text are processed on your device and are never uploaded to our servers.

Related tools