DjVu is the format of a great many scanned books: old textbooks, out-of-print titles, journals and the libraries of Internet Archive and of Russian and Eastern European sites, where it was the format of choice for years because it squeezes a scanned page much smaller than PDF can. The trouble is opening one. Windows, macOS, phones and e-readers have nothing built in for DjVu, the free DjVu readers are old, and the online converters upload your book and often cap its size.
This page reads the DjVu file directly and shows every page, drawn sharp at whatever size you zoom to. Most scanned books carry a hidden text layer, the text recognised from the scan, and the reader lays it over each page so you can search the whole book and select and copy words as if it were a PDF. When you want a file that opens anywhere, it writes a PDF: each page as an image, with the book's hidden text layer laid over it so the PDF can be searched and its text selected and copied, in any language, plus the book's bookmarks and links.
- How do I read a DjVu file online without installing anything?
- Drop it on the box above. The book opens in this browser tab on any computer, phone or tablet, with search, bookmarks and copying, and there is no account, no upload and no size limit beyond what your device's memory allows. Converting it is up to you: reading needs no PDF.
- Why can't I search this book?
- It has no text layer: it is scanned pictures of the pages with no recognised text stored in the file. Search says so when that is the case. The pages still show; reading the text from the pictures (OCR) is something this page does not do.
- Will the PDF be searchable?
- Yes, when the DjVu file has a text layer, which most scanned books from libraries and archives do: that text goes into the PDF, invisible over each page, so the PDF can be searched and copied from in any language. A DjVu file that is only scanned pictures has no text to carry over, and this page does not read the text from the pictures (OCR); the search box above says so when a book has none.
- Why is the PDF bigger than the DjVu file?
- DjVu was designed for scanned pages and stores them in layers, sharp text on top of a soft background, more compactly than PDF can. In the PDF each page becomes a picture with the black-and-white text kept sharp, which usually takes two to ten times the space. Smaller quality brings colour and grey pages down; black-and-white text pages stay much the same at every setting.
- Can I convert just some pages?
- Yes. Choose Pages and type the first and last page, such as 45 to 80 for one chapter. The book's bookmarks are kept only when all the pages are converted. Links from one page to another still work within the part, and a link to a page that is not in it is dropped rather than sent to the wrong page.
- My book is a folder of many .djvu files. Can it be converted?
- Not as a folder, for reading or converting. A DjVu book can be "indirect": an index file plus one file per page, as some sites serve them. This page reads the usual single-file (bundled) kind. DjVuLibre's free djvmcvt tool turns an indirect book into one bundled file, which then opens here.
- Is my DjVu book uploaded?
- No. The file is read and converted by this browser tab, with djvu-rs compiled to WebAssembly; nothing is sent anywhere. A very large book may take a minute or two to convert, and on a phone it can run out of memory; converting it in parts with the page range helps. Reading it is lighter: pages are drawn as you reach them and let go of when you move on.