Put a PDF into Claude as Markdown, not as an attachment

Claude reads PDFs on its own, so converting one first is a choice. Make it when you want to see and edit what the model receives, when a document is long enough to crowd the context window, or when the same text must serve every chat in a Project.

Drop your PDF or Word file here

or choose a file from your computer

PDF or Word (.docx) · up to 50 MB

How do I get a PDF into Claude?

Convert the PDF to Markdown at mypdf2md.com, then paste it into a Claude chat, add it to a Project's knowledge, or send it through the API as a plain-text document. Conversion runs in your browser, so nothing is uploaded.

Should I convert a PDF to Markdown or attach it to Claude?

Attach the PDF when charts and layout carry the argument, because Claude reads each page as an image as well as text. Convert to Markdown when the words are what matter: you can read, trim and correct what the model receives.

What does a PDF lose when it is converted for Claude?

Images are not extracted and equations arrive as plain characters rather than LaTeX. Headings, lists, tables, code blocks and emphasis survive as Markdown. A scanned page holds no text at all and needs the OCR converter instead.

How to put a PDF into Claude as Markdown

  1. 1

    Convert the PDF

    Drop the file onto the converter above. PDF.js parses it in a Web Worker on your own machine.

  2. 2

    Read what came out

    Check the Markdown against the preview: heading levels, table rows, and the reading order of anything in columns.

  3. 3

    Cut what you are not asking about

    Delete cover pages, revision histories and appendices. The counter under the editor shows what is left.

  4. 4

    Load it into Claude

    Copy it into a chat, or use Download .md for a Project's knowledge or an API request.

Why Markdown instead of the PDF itself

Claude does not read a PDF the way it reads text. Anthropic's PDF support documentation describes the process: every page becomes an image, with the text extracted from it supplied alongside. That is what lets Claude answer a question about a chart, and also why a long PDF is expensive: each page arrives twice over.

Markdown sends the words and nothing else. Headings stay headings, so the model can tell a section title from a short sentence. Tables stay tables rather than dissolving into loose numbers. Repeated page furniture is dropped. What structure buys a model makes the longer argument.

The trade-off runs the other way for some documents. Because this converter extracts text and not images, a figure leaves its caption behind and nothing else. Where charts carry the argument, attach the original PDF alongside the Markdown.

  • You read and edit the text before the model does
  • Repeated headers, footers and page numbers are removed automatically
  • Tables arrive as GitHub Flavored Markdown pipe tables
  • Conversion runs in a Web Worker on your machine — nothing is uploaded

Three places the output goes

In a chat, paste the Markdown straight into the message box. Claude reads Markdown natively, so it needs no wrapper. Put the document first and your question after it.

For a Project, use Download .md and add the file to the project knowledge, where every conversation there can draw on it. Anthropic's accepted-upload list covers plain-text documents; if an upload refuses .md, save it as .txt instead — Markdown is plain text, so nothing about the file changes.

Through the API the text goes in as a plain-text document rather than a PDF, skipping page rasterisation. For retrieval rather than a single paste, it has to be cut up first, and where you cut it decides what comes back.

Length, and what to do about it

A converted document is text, so its size is easy to judge: the editor counts words and characters as you trim. That measures what you are about to spend better than a page count.

Long documents do better split than whole. Anthropic's guidance for large PDFs is to split them into sections, and it survives the conversion: cut on the heading ladder, so each piece keeps a title saying what it holds. A section with its heading answers on its own; an arbitrary slice does not.

The free converter on this page reads the first 30 pages of a document. Password-protected files are accepted, and the password is used by the PDF reader in your browser and never sent anywhere, which matters because Anthropic's PDF path rejects encrypted files.

Where the conversion struggles

A scanned page is the clean failure: a picture of text and no text, so nothing can be extracted and the result comes back nearly empty. It needs optical character recognition, a separate converter here because it is the only one that uploads, with its own things it quietly drops.

Two-column papers are the interesting failure. The converter finds the vertical gutter and reads left before right, but a full-width table or wide equation crossing the middle defeats it, and the damage shows within a few lines. Putting the order back is quick.

Tables survive more often than not and still deserve a look, particularly where the original merged cells or broke one across a page; repairing the crooked ones usually takes a few pipe characters. Equations arrive as whatever characters the PDF stored rather than as LaTeX, so dense mathematics converts poorly.

Footnotes get no warning: they are ordinary text at the foot of a page, so they convert into ordinary paragraphs, landing between the sentence that cited them and what followed.

Frequently asked questions

Does Claude accept a .md file?
Anthropic's accepted-upload list for chats and projects covers plain-text documents. Markdown is plain text, so if an upload refuses .md, saving the file as .txt changes nothing about its contents.
How many pages does the free converter handle?
It converts the first 30 pages of each document, and says so when a document runs longer. There is no limit on how many documents you convert, and no account is required.
Is anything sent to Anthropic while the PDF is converted?
No. The conversion happens in your browser, and no request carries your document to Anthropic, to us, or to anyone else. What you do with the Markdown afterwards is a separate step.
My scanned PDF converted to almost nothing. Why?
A scan is a picture of a page, so it carries no text layer and there is nothing to read. Use the scanned PDF converter instead: it recognises the text on our server, says so before sending anything, and is the only page here that uploads.

Related converters

Further reading