PDF to EPUB Converter

How to convert a PDF to EPUB

1. Choose the PDF. Drop it on the box above or pick it from your device. The page reads the file in your browser; nothing is uploaded.
2. Check the details. Title and author come from the PDF and can be edited. The page also shows the page count, whether the PDF has a text layer and whether it has bookmarks.
3. Pick your reader. Choose the preset closest to your screen. It sets the image size, margins and line spacing written into the EPUB.
4. Read the preview. The first pages of the finished book appear beside the controls. Change the preset or the title and the preview and the file update.
5. Download the EPUB. Copy it to your reader over USB or send it with your reader’s email service.

The converter is built for books and long articles that have a real text layer. If you can select and copy text in a PDF viewer, the PDF has one. Pictures of pages (scans) need text recognition first, which this site does not offer yet; see the section on scanned PDFs below.

What the converter does to a PDF

A PDF stores where each piece of text sits on a page, not what a paragraph or a chapter is. Turning it into a book means rebuilding that structure. Here is what is rebuilt, and how.

• Reading order. Text is grouped into lines and paragraphs. On pages with two columns the first column is read before the second, and a full-width title stays on top. Rotated side notes are left out.
• Page furniture. Running headers, footers and page numbers are text that repeats at the same height on many pages. They are detected and removed, so “12” and the book title no longer appear in the middle of a sentence.
• Paragraphs and hyphens. Lines are joined into paragraphs, including paragraphs that continue on the next page. A hyphen at the end of a line is removed when the joined word is a real word, and kept in compounds such as well-known or mother-in-law. Ligatures such as fi and fl become plain letters.
• Chapters. If the PDF has bookmarks, they decide the chapters and their titles. Otherwise headings are found by size, weight and wording such as “Chapter 3”. Every chapter becomes its own file and its own line in the table of contents, in both the EPUB 3 and the older NCX form.
• Images. Pictures stay where they were in the text and are scaled to the preset. Tiny ornaments, rules and full-page backgrounds are skipped. Photos are saved as JPEG; line art stays PNG when that is smaller.
• Styling. Italic and bold are kept. The stylesheet is plain: no fixed positions, no fixed page size, so the reader’s own font and size settings apply.

Reader presets and what they change

A preset is a short list of numbers, shown below the controls before you convert. It does not change the text, only how images are saved and how much room the page leaves around the words.

Greyscale images are smaller and look right on e-ink. You can switch them off for colour screens.

Make a PDF reflowable on Kindle or Kobo

A PDF is a fixed page, so a 6-inch screen shows it tiny or makes you pan around. An EPUB reflows: the reader sets the font, the size and the line breaks. Use the converter above, then pick the preset by screen size. Typically, a 6-inch Kindle or a Kobo Clara wants the 6-inch preset; a 6.8-inch Kindle Paperwhite, a 7-inch Kindle Oasis or a Kobo Libra suits 6.8–7 inch. If you are not sure, look up your reader’s screen size.

1. Kobo. Kobo readers open EPUB directly. Connect the reader with USB and copy the file into its storage, or add it through Kobo’s desktop app.
2. Kindle. Kindle readers take EPUB through Amazon’s Send to Kindle (website, app or email), which converts it for the device. Check Amazon’s page for the formats it currently accepts.
3. Other readers. Small e-ink readers and reading apps on phones open an EPUB copied into their books folder.

Text PDFs, scanned PDFs and mixed PDFs

Text PDFs are converted here, now, for free. Scanned PDFs are photographs of pages. They contain no text, so there is nothing to rebuild until text recognition (OCR) has read the pictures. This site does not do that yet; it tells you when it finds a scan and does not pretend otherwise. Mixed PDFs have some text pages and some picture pages. The text pages are converted, the picture pages are kept as images in the right place, and you are told how many there were.

We are preparing a scanned-book plan (Scanned book, $2.99 per book). It is not open and nothing is charged. If you want it, you can reserve a place; there is no date.

Free limits

Conversion is free for text PDFs up to 400 pages and 100 MB. The limit exists because the whole job runs inside your browser tab: every page is parsed, laid out again and its images are re-encoded using your device’s memory and processor. 400 pages is a long novel with room to spare, and it keeps phones and older laptops from running out of memory. A bigger PDF is not refused: the first 400 pages are converted for free, and you are told what was left out.

Where the converter struggles

• Tables, forms and maths. They are text on a grid, not paragraphs. Expect the cells to run together.
• Footnotes and sidebars. They are placed where they sit in reading order, which may be in the middle of a paragraph.
• Three or more columns, magazines, slides. The layout is too free for a book; the result is readable but not tidy.
• Right-to-left and vertical scripts. Line grouping assumes horizontal text.
• Odd fonts. If a PDF’s fonts map letters to the wrong characters, the text layer itself is wrong, and so is the EPUB.

If a result looks wrong, try the other preset, check whether the PDF has bookmarks, and read the FAQ on messy results below.