Full issue
- Input
- Terrors of the Jungle #10, a 25,006,136-byte CBZ of 36 JPEG pages.
- Output
- 25,901,124-byte EPUB in 1.05 s; EPUBCheck 5.2.1 reported no errors or warnings.
CBZ → EPUB wraps each comic page in its own fixed page, so reading apps that shelve EPUB but ignore CBZ show the comic with a cover in the library.
From real runs
| Element | Result | What happens |
|---|---|---|
| Fixed pages | Kept | The EPUB is pre-paginated: every page has a viewport the size of its scan, so a reading app shows whole pages and never reflows them. |
| JPEG and PNG scans | Kept | Scans are stored in the EPUB unchanged; only the file names become p0001.jpg, p0002.png and so on. |
| WebP pages | Changed | WebP pages are saved as JPEG for older reading apps, which is why the mixed test archive grew from 5.36 MB to 5.71 MB as EPUB. |
| Cover and title | Kept | The first page is marked as the cover image for library views, and the title comes from ComicInfo.xml or, failing that, the file name. |
| Right-to-left manga | Kept | When ComicInfo.xml says Manga YesAndRightToLeft, the EPUB is marked right-to-left so page turns go the Japanese way. |
| Text size and reflow | Lost | A scanned page has no text to enlarge or reflow; you zoom the picture as you would a PDF. |
| Table of contents | Changed | The contents list has a single entry pointing at the cover, because a CBZ carries no chapter information. |
The EPUB is pre-paginated: every page has a viewport the size of its scan, so a reading app shows whole pages and never reflows them.
Scans are stored in the EPUB unchanged; only the file names become p0001.jpg, p0002.png and so on.
WebP pages are saved as JPEG for older reading apps, which is why the mixed test archive grew from 5.36 MB to 5.71 MB as EPUB.
The first page is marked as the cover image for library views, and the title comes from ComicInfo.xml or, failing that, the file name.
When ComicInfo.xml says Manga YesAndRightToLeft, the EPUB is marked right-to-left so page turns go the Japanese way.
A scanned page has no text to enlarge or reflow; you zoom the picture as you would a PDF.
The contents list has a single entry pointing at the cover, because a CBZ carries no chapter information.
We converted our Comic Book ZIP (CBZ) sample (325 KB) with Image engine on 2026-10-20. It took 18 ms and produced a 327 KB EPUB file, about the same size as the input. The engine raised no warnings.
Input · sample.cbz · 325 KB
ZIP with 4 parts
Output · sample.epub · 327 KB
EPUB 3.0: 4 content documents, 3 images, EPUB 3 nav, cover set, title “Bobby Make-Believe”
calibre
ebook-convert comic.cbz comic.epubcalibre 6.13 shrank 975 × 1349 scans to 584 × 754, stored them as PNG, added its own title page and skipped WebP pages; check its comic input options before converting a collection.
The HTTP API and the MCP server run the exact engine this page uses (Image engine); the response carries the same warnings the web converter shows. Check support first with GET /api/v1/pairs?from=cbz&to=epub.
curl -X POST https://filetypeconverters.com/api/v1/convert \
-H "Authorization: Bearer $FTC_API_KEY" \
-F "file=@input.cbz" -F "from=cbz" -F "to=epub" \
--output output.epubNo. Comic pages are pictures, so the EPUB uses fixed layout: one scan per screen, zoomed the way your reading app zooms pages.
The page images are stored as they were, and the EPUB adds only small XHTML wrappers. It grows only when WebP pages have to be saved as JPEG.
Any reading system that supports EPUB 3 fixed layout. Our files pass EPUBCheck with no errors; we have not tried every e-reader, so open one issue before converting a whole series.