PDF to Word, free
Turn a PDF into an editable .docx — keeping the text, headings, lists and pictures. The document stays in your browser; nothing is uploaded.
Drop the PDF here, or choose a file
The document stays in your browser — nothing is uploaded.
You can drop several PDFs at once — up to ten — and download the finished Word files in one ZIP.
Why in the browser?
What you want back in Word is typically a contract, a CV, a thesis or an official letter you need to change. Most "PDF to Word" sites upload it to a server, convert it there and keep it for days — your document travels through a stranger's machine. This page does not work that way: your browser reads the PDF, your browser writes the Word file, and not one byte leaves your machine.
The engine is two open-source libraries: pdf.js, which Firefox also uses to display PDFs, reads the text and the pictures together with their positions; docx.js writes the result in Word's own format. There is no model to download, and even the first run takes a few seconds.
What survives, and what does not
A PDF has no paragraphs — only letters, with coordinates. What belongs together is inferred from where things sit: the distance between lines, the indentation, the font size. From that come paragraphs, headings (three levels), bullet and numbered lists, centred or right-aligned lines, and bold and italic runs in Word. Pictures go into the text at their drawn size, in the right place, and every PDF page ends with a page break so the pagination survives.
What cannot be rebuilt: tables (their cells arrive as text, row by row), multi-column layouts (the columns follow one another), text boxes, form fields, footnotes, and headers and footers as separate elements. The original font name does not travel either; Word uses its default typeface. That is "layout-approximate": the text and its structure come through, not the pixel-exact picture.
For a scanned PDF, OCR first
If the PDF is a scanned sheet or a photographed page, there is no text in it — only a picture of text. This tool then finds nothing to convert, and says so instead of handing you an empty document. The answer is the text recogniser (OCR): it reads the letters off the picture and produces a PDF whose text is searchable. Drop that one here and it becomes a Word file.
Frequently asked questions
Will it look the same in Word as the PDF?
No, and no converter can honestly promise that. What comes through: the text, paragraphs, headings, lists, bold and italic runs, pictures and page boundaries. What does not: table structure, columns, text boxes, the original typeface. You get an editable document, not a pixel-exact copy of the PDF — which is the point if you want to change it.
Can it convert a password-protected PDF?
Yes, if you know the open password: the page asks for it, and your browser opens the file with it. The password never leaves your machine and is not stored. In a multi-file batch a protected file is skipped — remove its password first with the password remover.
Why is the document empty, or why is text missing?
Because the PDF held a picture rather than text — a scanned or photographed page. The page tells you before the conversion and points at the text recogniser. If only some pages are like that, the result shows how many pages had no text.
How large a PDF can it handle?
The limit is the memory on your machine: a document of several hundred pages full of pictures goes through on a laptop; on a phone, files over 100 MB may get slow. Up to ten files are processed at once, and pictures are reduced to at most 2000 pixels wide so the Word file does not become huge.
Does it open in Google Docs and LibreOffice too?
Yes. We write a standard .docx with built-in heading styles and real list numbering, so Word, Google Docs, LibreOffice and Pages all open it the same way — the navigation pane and a table of contents work from it.