Convert PDF to Markdown
Turn a PDF into clean Markdown. Headings come from the type sizes, lists and paragraphs are put back together, and page numbers and running headers go. Use it in notes apps and AI chats. It all runs in your browser.
- Free, no sign-up
- Runs in your browser. Your files never leave your device.
- Updated
or paste a file with Ctrl V
- Words
- 0
- Pages
- 0
- Characters
- 0
Read them with Image to Text, which runs OCR on PDFs, also in your browser.
Reading pages
To use it in Gemini Notebook (formerly NotebookLM), add a source, choose Copied text and paste. Gemini Notebook also takes the PDF itself, up to 500,000 words or 200 MB per source.
Kortex PDF to Markdown converts a PDF’s text layer into Markdown in your browser. Larger type becomes # headings. Bullets and numbers become lists. Broken lines and hyphenated words are rejoined. Page numbers and repeated headers are removed. Download one .md file or one per page. Nothing is uploaded.
How do I convert a PDF to Markdown?
- Drop the PDF on the box above. Reading starts at once.
- Check the Markdown in the preview. Turn on Page markers if you want page numbers kept as comments.
- Click Download .md, or Copy to paste it into a note or a chat.
Why Markdown instead of plain text?
Markdown keeps the outline. In Obsidian or Notion, the headings become an outline you can click through, and the lists stay lists. AI assistants such as Gemini Notebook, ChatGPT and Claude can tell a section title from the sentence under it. Plain text flattens all of that into paragraphs.
Getting the best Markdown from a PDF
- PDFs made from Word, Google Docs or LaTeX convert best: they have clear heading sizes and real text.
- If a heading comes out as a paragraph, it was set at body size (often just bold). Add the
#by hand. - Select pages in the Pages panel to convert one chapter at a time.
- Scans need text recognition first. Use Image to Text.
Need plain text instead? Switch to Text above, or use PDF to Text.
Questions
How does it know what is a heading?
By type size. The size most of the text uses counts as body text. Lines set clearly larger become headings: the largest as # and smaller ones as ## and ###. Bold text at body size stays a paragraph.
Are lists and paragraphs kept?
Yes. Lines that start with a bullet (•, ●, –) become - items, and numbered lines become 1. items. Lines of a paragraph are joined into one. Words broken with a hyphen at a line end are put back together.
What about tables and images?
Tables come out as lines of text, not Markdown tables. A PDF stores a table as text placed in a grid, not as rows and cells. Images are left out.
Can I mark where each page starts?
Turn on Page markers to put <!-- Page 3 --> before each page’s text. It does not show when the Markdown is rendered, but you can search for it. It also lets an AI see which page a passage came from.
Is the PDF uploaded?
No. pdf.js reads it in your browser tab, and the .md file is made on your device.
What if the PDF is a scan?
Scanned pages have no text to convert. The tool lists them. Image to Text can read them with OCR (text recognition) in your browser.
More free tools
- PDF to Text Get clean text or Markdown out of any PDF, page by page.
- Merge PDF Join PDFs into one file. Reorder and rotate pages in your browser.
- Split PDF Split a PDF by page, range or size. Or pull out the pages you need.
- Image to Text (OCR) Copy the text from photos, scans and screenshots. 100+ languages.
- Background Remover Remove the background from any photo. The AI runs on your device.
Send it to your notebook in one click.
Kortex sends web pages, highlights, posts, YouTube videos and AI chats into the Gemini Notebook you pick. Bulk import also takes your open tabs, CSV files and RSS feeds (lists of a site’s new posts).
No card needed.