iloveepubProcessed locally — files never leave your device
PDF to EPUB
Turn a PDF book into a reflowable EPUB you can resize on any reader — with automatic scan detection before anything is converted. Free, no upload, no limits.
01Select your PDF
How to convert a PDF to EPUB
Select your PDF file
The file is analyzed locally in your browser — a scan check runs in milliseconds before anything is converted.
Confirm the book title
We pre-fill the title detected in the PDF. Edit it if you like; it is used for the EPUB metadata and contents.
Download the EPUB
Chapters are detected from headings, typeset cleanly and assembled into a reflowable EPUB saved to your device.
Reflowable vs fixed: why anyone converts at all
PDF and EPUB embody two philosophies. A PDF is a print master: pages of a fixed size, every glyph pinned to a coordinate, beautiful on paper and on screens large enough to show a full page. An EPUB is a web page in book clothing: text that re-wraps to any width, a font size the reader controls, and layout that adapts from a 6-inch e-reader to a 13-inch tablet.
Converting PDF to EPUB means un-fixing the page. Done well, you gain everything e-readers are good at — instant font-size changes, dictionary lookup, read-aloud, progress syncing, and comfortable one-handed reading on small screens. Done poorly, you gain a mess. The difference is almost entirely explained by one question: what kind of PDF are you starting from?
The two kinds of PDF — and only one converts
Open your PDF and try to select a sentence. If a highlight appears over the letters, you have a text-based PDF: the file contains actual characters with font and position information. Publishers export these directly from typesetting tools, and they are the raw material a converter can work with.
If instead your cursor drags a box over a picture of the page, you have a scan — every page is a photograph of paper. There is no text inside to extract, only ink. Recovering text from a scan requires OCR (optical character recognition), which is a heavyweight, error-prone process that cannot honestly be done in a browser tab. Every honest converter — ours included — will refuse scans rather than hand you an empty or garbled book. A third variant, a scanned PDF with an invisible OCR text layer, sits in between: it can convert, but inherits the OCR's mistakes.
What the conversion pipeline actually does
A text-based PDF stores lines, not paragraphs: each line is a run of characters at a position. The converter's first job is reconstruction — grouping lines into paragraphs, rejoining words broken by hyphenation at line ends, and discarding running headers, footers and page numbers that would otherwise sprinkle the text.
The second job is structure: spotting chapter titles from font size and spacing, building a table of contents from them, and marking up headings, emphasis and block quotes. The third is assets — extracting embedded images and re-embedding them in the EPUB at sensible sizes. Get all three right and the output is indistinguishable from a native EPUB; get paragraph reconstruction wrong and sentences arrive interleaved, which is the classic failure of cheap converters.
What you lose, what you gain
Honest ledger. You lose exact layout: multi-column magazine pages will reflow into single columns in reading order (usually right, occasionally wrong), and designer typography becomes reader typography. Annotations you made in the PDF stay in the PDF; they belong to the viewer, not the file.
You gain the e-reader's whole toolkit: text that fits the screen you're holding, fonts you can grow tired eyes into, a chapter list that jumps anywhere, and a file that weighs a fraction of the original. For books — as opposed to forms, sheet music and slide decks — it is almost always a good trade.
How to check your PDF before converting
Three quick tests. The selection test above: selectable text means convertible. The size test: a 200-page book at 60 MB is almost certainly scans — images are heavy, text is cheap. The title-page test: zoom in on a page corner; scanned pages blur like photographs, while born-digital text stays razor sharp at any zoom.
Our converter runs this diagnosis for you automatically, in milliseconds, before any conversion is attempted — a scanned book gets a clear explanation instead of a broken output file.
PDF types and how they fare in conversion to EPUB
| PDF type | What's inside | Converts to EPUB? |
|---|---|---|
| Text-based | Real characters with font + position data | Yes — the good case |
| Scanned | A photo of each page; no text layer | No — needs OCR, not possible in-browser |
| Scanned + OCR layer | Photo plus hidden, imperfect text | Partially — inherits OCR errors |
| Hybrid | Some chapters text, some scans | Text chapters only |
| Forms / tagged | Interactive fields or accessibility tags | Fields drop out; tags can help |
Frequently asked questions
Why won't you convert scanned PDFs?
Because it cannot be done honestly in a browser. Scans contain pictures of text, and recovering that text requires OCR — heavy, error-prone computation that we cannot run locally. Rather than ship a tool that silently produces garbage, we detect scans and tell you up front.
Is the PDF to EPUB converter free?
Yes — like every tool on this site: no accounts, no credits, no page-per-day limits.
Will my PDF be uploaded anywhere?
No. Parsing, diagnosis and conversion all run in your browser. The privacy promise on this site is structural, not a policy: there is no upload endpoint at all.
How does the tool decide where chapters start?
It looks at the headings recovered from the PDF: if the book has clear top-level headings, each becomes a chapter in the EPUB's table of contents. Books without heading structure come through as one continuous, well-typeset file.
Can I go the other way, EPUB to PDF?
Yes — that direction is simpler because typesetting a book is easier than un-typesetting one. See the EPUB to PDF page for how it works.