Made for scanned books and mixed page batches
Use this tool when a scanner captures two book pages in one image, when sideways scans interrupt a document, or when a folder contains covers, blanks and loose pages in the wrong order. All split positions and crops are manual, so you stay in control of illustrations, notes and narrow margins.
It does not recognize text, guess page boundaries, flatten curved book pages, deskew small angles or correct perspective. For searchable text, run OCR separately after you finish the page layout. For sharply tilted phone photographs, correct perspective before importing.
Practical limits and export trade-offs
A batch that would exceed a limit is rejected with a message, rather than silently losing pages. The workspace supports 100 source pages and 200 output pages, input files up to 50 MiB each and 150 MiB in total, and still images up to 30 MP. Very large or malformed PDFs may still exceed your browser’s available memory. PDFs with inline images, unsupported content compression or predictors are rejected; convert those pages to PNG or JPEG first.
Export uses one JPEG raster per included page. Pages are rendered sequentially and capped at 12 MP and 4096 pixels on either side; the finished file has a 150 MiB size limit. If it is too large, lower the DPI or omit pages and export smaller groups. Page count is checked before a download is offered.
What stays local, and what uses the network?
Imported file bytes, page pixels, file names, crop values and generated PDFs are processed in this browser session. There is no document-upload endpoint and the workspace is not saved to an account. Reloading, clearing or leaving can discard your work; switching language opens a separate page.
PDF rendering and PDF export use pinned PDF.js 6.4.299 and pdf-lib 1.17.1 libraries loaded from jsDelivr, including PDF worker and font resources. Loading the page and those resources still makes network requests. Google Analytics is configured with the canonical page URL for visit statistics; this tool does not send file names, contents or edit values to analytics. The page is not guaranteed to work fully offline.