PDF to Markdown
Turn a PDF into Markdown for a language model or a notes app. Headings are guessed from font size and lists from bullets; nothing is uploaded.
This tool does not upload your files. Close the tab and nothing is left behind.
Getting a PDF into a text pipeline
Feeding a PDF to a language model, a search index or a notes app means getting the words out first. A ninety-page contract is not something you paste into a chat box, and copying it page by page from a viewer loses track of where each passage came from. Markdown with page markers keeps the text usable and keeps the page numbers attached to it.
This tool reads the text layer the PDF already contains, groups the fragments into lines by their position on the page, and writes each page as its own block with a comment naming the page number and a divider after it. Structure is guessed from the layout: lines set clearly larger than the body text become headings, bulleted and numbered lines become list items and wrapped lines are joined into paragraphs, while bold text and tables are not detected. Multi-column layouts can interleave, and a scanned PDF returns nothing at all because its pages are pictures, so run those through the OCR tool first.
The conversion happens in your browser: the file is read into memory on your own machine and never sent anywhere, which is the whole point when the document is a contract, a medical record or anything else you would not paste into a stranger's server. Because every page keeps its number in a comment, a passage that came out wrong is easy to find and fix later. If you need real tables and headings rather than plain prose, the AI conversion rebuilds them, but that one does send your pages to Google, which is a different bargain.
How to use
- 1
Select a PDF.
- 2
Optionally set a page range.
- 3
Click Start and download the .md.
Frequently asked questions
- How is this different from AI conversion?
- This reads the text the PDF already stores, entirely in your browser. AI conversion rebuilds tables and headings with your own Gemini key, which means the pages are sent to Google.
- What is the page marker format?
- Each page starts with a comment such as <!-- Page 3 --> and ends with a --- divider.
- Are headings and lists detected?
- Yes, from the layout: lines set larger than the body text become headings and bulleted or numbered lines become list items. A complex document can come out with odd heading levels; AI conversion is more accurate there.
- Are tables and images converted?
- No. Tables are not recognised, so cell text comes out as plain lines, and images are left out. Use AI conversion for tables and Extract images for the pictures.