How to Convert PDF to EPUB (Reflowable E-Book, Free)
Turn a PDF into a reflowable EPUB for Kindle, Kobo or your phone. Works best for text-heavy PDFs — here's what converts cleanly and what degrades, honestly.
Reading a PDF on an e-reader or a phone means endless pinch-zooming: the page is wider than your screen, the type is tiny, and every line needs a horizontal scroll. The fix is to convert the PDF into EPUB — the native format of Kobo, Apple Books, Google Play Books and (since Amazon accepts EPUB via Send to Kindle) Kindle too.
EPUB text reflows: it wraps to your screen width, at the font size you pick, with night mode and dictionary lookup working the way the device intends. For a 300-page text PDF, the difference in reading comfort is enormous.
How to convert PDF to EPUB — step by step
Open the tool
Go to pdfty.com/tools/pdf-to-epub. Browser-based — nothing to install.
Upload the PDF
Drag your file in. Free up to 20 MB and 50 pages, 10 operations per day.
Convert
Text, headings and images are extracted from the fixed pages and rebuilt as flowing chapters. Takes seconds for most documents.
Download the EPUB
A standard .epub any e-reader app opens.
Send it to your device
Email it to your Send-to-Kindle address, drop it into Kobo/Apple Books, or open it in any reader app. Server copies are deleted within 1 hour.
What converts well — and what degrades
Here's the honest physics of the problem: a PDF stores characters at fixed coordinates on fixed pages. To make text reflow, the converter must throw that positioning away and reconstruct the logical reading order. For linear text that's reliable. For visually complex pages it can't be — the "layout" and the "flow" are fundamentally at odds.
| PDF content | Result in EPUB | Verdict |
|---|---|---|
| Novels, reports, essays — straight text | Fixed pages | Clean flowing chapters |
| Headings and emphasis | Fixed pages | Preserved as styled text |
| Inline images and figures | Fixed pages | Carried over, placed in the text flow |
| Simple tables | Fixed pages | Usable, may wrap awkwardly on small screens |
| Multi-column layouts (papers, magazines) | Fixed pages | Reading order usually right, visual layout gone |
| Math, code blocks, precise diagrams | Fixed pages | Often mangled or flattened — keep the PDF |
| Scanned pages without OCR | Fixed pages | No text found — OCR first |
Free local alternatives
- Calibre. The desktop standard:
ebook-convert input.pdf output.epub, or use the GUI. Its PDF input has the same inherent limits (PDF reflow is hard for everyone), but it offers knobs — un-hyphenation, heading detection — worth it if you convert often. - Send to Kindle without converting. Amazon accepts PDFs directly and can convert them to reflowable format on their side. Quality is comparable; you're limited to the Kindle ecosystem.
- Copy into a document, export EPUB. For short texts: paste the content into a word processor and export EPUB (LibreOffice and Pages can). Manual, but you control the cleanup.
The browser tool is the quickest path for one-off conversions, on any device, with nothing installed.
Practical gotchas
Page headers and footers. Running headers, footers and page numbers from the PDF are detected and dropped where possible — in a reflowing book they'd be noise mid-paragraph. Occasionally one slips through; a quick edit in Calibre fixes stragglers.
Hyphenation. PDFs break words at line ends. The converter rejoins them, but rare false joins ("sidebar" vs "side-bar") can occur in unusual typography.
Chapter detection. Chapters are inferred from headings and page structure. Books with clear heading styles get a proper table of contents; documents without visible structure become one long flow.
Scans need OCR first. If your PDF is a scanned book, run it through OCR before converting — our free OCR guide explains accuracy. Reflowed OCR text inherits any recognition errors, so proof-skim the result.
Images on e-ink. Color figures render grayscale on e-ink readers, and very wide images get scaled down. That's the device, not the file.
Frequently asked questions
Will the EPUB look like the original PDF?
No — by design. EPUB discards the fixed layout so text can adapt to your screen and font settings. Content, order, headings and images are preserved; page geometry is not. If you need the exact appearance, stay with the PDF.
Does it work for Kindle?
Yes. Amazon's Send to Kindle accepts EPUB (it converts to its internal format on delivery). Email the converted file to your @kindle.com address or use the Send to Kindle app.
My PDF is a scanned book. Can I convert it?
Not directly — a scan has no text layer. Run OCR first to recognise the text, then convert the OCR'd PDF to EPUB.
Why does my two-column paper read strangely?
Two-column PDFs are the hardest case: the converter must guess column boundaries to get reading order right. It usually succeeds, but captions, sidebars and footnotes can land mid-text. Academic papers with equations are better read as PDF.
What are the limits, and what does it cost?
Free: 20 MB, 50 pages, 10 operations per day — no sign-up, no watermark. Pro is $9/month for unlimited use. Uploads and outputs are deleted from the server within 1 hour.
Can I edit the EPUB afterwards?
Yes — Calibre's editor or Sigil (both free) let you fix headings, split chapters or delete a stray header line. Small cleanups there can noticeably polish an imperfect conversion.
Is EPUB better than PDF for reading?
On small screens and e-readers: yes, decisively — reflow, your fonts, night mode. On paper or for layout-critical documents: no — that's PDF's home turf. The two formats coexist for a reason; convert back anytime with EPUB to PDF.
Will links and the table of contents work?
Detected headings become EPUB navigation entries, so you get a working TOC in your reader for well-structured documents. Web links in the text generally remain tappable.
L'équipe pdfty crée des outils PDF en ligne respectueux de la vie privée — compresser, convertir, OCR, signer et protéger. Fichiers supprimés sous 1 heure. À propos →