Combine several PDFs in whatever order you arrange them.Pages are moved across as-is — no re-rendering and no re-compression, so sharpness is exactly the same.
Does merging reduce quality?
No. Pages are copied as-is: no re-rendering and no re-compression, so quality matches the originals exactly.
Can it merge encrypted PDFs?
It tries with encryption ignored. If some pages come out blank or garbled the file needs a password — unlock it first.
Everything runs locally in your browser — files are never uploaded. BoxKit VIP unlocks unlimited use of all 29 tools.
You have several PDFs to hand to the same person: medical reports, a dozen invoices, a weekly lab record, or a tender pack of business licence plus certificates plus quotation. What they share is that page order matters, the content cannot change, and the other side wants a single file.
The same applies to double-sided printing or binding. A printer handles one file best for inserting blank pages and paginating automatically; drop five files into the print queue separately, with another job cutting in, and they come out in the wrong order.
Check two things before dropping files in. A file that needs a password to open cannot be read, so those pages come out blank — save an unencrypted copy first. Files that only restrict printing or copying can be merged, but those restrictions are not carried over. A file that cannot be read at all stops with an error instead of producing half a document; for a readable file with a few broken pages, re-saving it via Print to PDF in its original reader usually fixes it.
Desktop software is clearly better at some things. Acrobat's Combine Files can turn each file's table of contents into new bookmarks and re-stamp continuous page numbers; macOS Preview merges by dragging thumbnails. This tool does neither, so use them when that matters to you.
Command-line tools take a different route. Ghostscript reinterprets the whole document and re-emits it, which can shrink it as a side effect — at the cost of rewriting images with new compression settings and possibly changing font encoding, so scans often come out blurrier. This tool does not rewrite page content, so it cannot reduce size; use PDF compression for that.
The difference with online merging sites is where the file goes. Upload a contract, medical record or ID scan once and it has left your device, and many sites also cap single files at 20–50MB. This tool reads and writes locally with pdf-lib — the file never goes through a server, and once the page is open you can finish merging offline. The trade-off is that the result lives only in this tab's memory: refresh or close the page and it is gone.
The pages are moved over — what exactly is moved?
A page in a PDF is not an image but a tree of objects: the page dictionary records its size and rotation and references fonts and image streams. Merging copies those objects verbatim into the new file's object table — the bytes that decide what the page draws are not regenerated, and embedded images are not decoded and re-encoded. Text stays text, lines stay vector, and scanned pixels are untouched.
After reordering, how do the page numbers map?
Page 1 of the second file in the list becomes page (pages in the first file + 1) in the result. Move a file up one place and its starting page moves back by the page count of the file above it. The page count shown after each file is there for exactly this calculation, so fix the order before merging.
Why is the merged file size almost the sum of the individual files?
Every page's content stream and embedded images must be stored in full; duplicate content is not deduplicated, so dropping the same file twice uses twice the space. Slightly smaller happens because an uncompressed cross-reference table or trailing junk in the original is dropped on rewrite, usually saving tens of KB up to about 1%; slightly larger happens because each source keeps its own resource tables and structure. To actually shrink it you need re-compression.
Do the header and Page X of Y change along with it?
No. Those page numbers were drawn onto the page when the file was created and are hard-coded into each page's content stream — they are not fields that recalculate themselves. A cover reading "Page 1 of 12" may now be page 15 of the merged file and still shows 1/12. For continuous numbering you must add it afterwards with a tool that supports it.