PDF to Word: convert PDF to editable DOCX online free
Free in-browser PDF to Word converter. Extracts text from every PDF page, groups it into paragraphs and writes a fresh .docx that opens in Word, Google Docs and LibreOffice. Powered by PDF.js + the docx library — no upload.
Drop a PDF to convert
Or click to choose. One file at a time. Stays on your device.
Choose PDFImage mode renders every PDF page as a high-resolution JPEG sized to match the source page and writes it into the DOCX with zero margins — the result looks identical to the original PDF when opened in Word, but the text isn't selectable / editable. Editable mode extracts text with per-run font size, bold, italic, family, detects headings and tables. Editable means you can rewrite the content, but the visual layout won't match the source. There is no free open-source library that gives both at once; Smallpdf, iLovePDF and Adobe use commercial server-side engines (Solid Documents, ABBYY).
When to convert PDF to Word
Edit a contract
Open a received PDF as DOCX and edit clauses without retyping.
Quote & paste
Pull plain text out of PDFs into emails, blog posts or notes.
Search & replace
Update product names or pricing across many PDFs by editing them as Word docs.
Translate
Paste extracted text into your translator of choice without scraping the PDF.
How to convert PDF to Word
Drop the PDF
Drag a PDF onto the page, click to browse, or paste from clipboard. PDF.js parses the document locally — nothing is uploaded.
Pick the output mode
Image-based (default) embeds each PDF page as a JPG inside the DOCX — perfect for keeping the exact layout. Text-based extracts editable paragraphs with bold / italic / font size preserved per run.
Click Convert
For text mode, paragraphs are inferred from Y-position changes, runs from font-size and weight changes, and tables from column clustering. For image mode, each page is rendered at high DPI and embedded.
Download the .docx
Open in Microsoft Word, Google Docs, LibreOffice or Apple Pages — every modern editor reads the standard DOCX format we produce.
About converting PDF to Word in the browser
Converting PDF to Word in a browser is one of the hardest format-translation problems in document software. PDF is a fixed-layout format that describes "draw this glyph at this exact coordinate"; Word is a reflowable format with paragraphs, runs, headings and styles. Going backwards from fixed-layout to reflowable requires guessing the original document model from page coordinates — which is exactly what commercial engines like Solid Documents and ABBYY do, and which no free open-source library matches in fidelity. That's why this tool gives you two modes.
Image-based DOCX (default)
Each page is rendered with PDF.js at high DPI, exported as a JPG, and embedded into the DOCX wrapped in a section sized to the original PDF page (1 PDF point = 20 twips). Margins are zeroed so the page looks pixel-identical to the original. The output is not editable text but the layout is perfect — best for forms, designed documents, brochures and any PDF where appearance matters more than editability.
Text-based DOCX
For text-heavy PDFs (reports, papers, contracts, articles) the text-extraction mode reads every glyph's position, font family, size and weight from PDF.js, clusters them into paragraphs by Y-coordinate gaps, splits each paragraph into runs by font-size / bold / italic changes, and emits the result as styled DOCX paragraphs. Tables are detected by column clustering — runs of text that share consistent X-coordinates across multiple lines become a real <w:tbl>. The result is fully editable in Word.
What's preserved, what isn't
Text mode preserves text content, paragraph breaks, per-run bold / italic, font size and basic table structure. It does not preserve exact font matching (Word substitutes if the original font isn't installed), multi-column layouts (collapsed to single column), inline images, footnotes, headers / footers, or the original page background. Image mode preserves everything visually but nothing is selectable text.
Scanned PDFs need OCR first
A scanned PDF is a sequence of page images with no real text data — there's nothing for the text extractor to pull. Run it through our PDF OCR tool first to add a selectable text layer (powered by Tesseract.js), then convert.
Privacy
The entire pipeline runs in JavaScript on your device. PDF.js parses the input; the docx library writes the output. Nothing is uploaded, nothing is logged. You can use the tool offline once the page is loaded.
When to pick a different tool
If you only need the text and not the formatting, copy-paste from any PDF reader is faster. If you need a pixel-perfect copy to view rather than edit, use our PDF to Image or PDF to PowerPoint tools. If the PDF has complex multi-column scientific layout, Adobe Acrobat Pro is still ahead of any free tool.
PDF to Word — frequently asked questions
Drop the PDF in, optionally pick a page range, then click Convert. Each page's text is extracted with PDF.js, grouped into paragraphs and written into a fresh .docx using the docx library. The output is editable in Word, Google Docs and LibreOffice.
Text content is preserved. Paragraph breaks are inferred from y-position changes. Complex multi-column layouts, exact font matching and overlapping graphics are not retained — PDF and Word are very different formats. For a pixel-perfect copy that's not editable, use our PDF to PowerPoint or PDF to Image tools.
No. PDF.js parses the file and the docx library builds the DOCX entirely inside your browser. The PDF never leaves your device.
Scanned PDFs are images of text and contain no real text data. Run them through our PDF OCR tool first to extract a text layer, then convert with this tool.