Merge PDF

Combine multiple PDFs into one, entirely in your browser

Merge PDF joins two or more documents into one, in the order you arrange them. It creates an empty document with pdf-lib and copies the page objects from each source into it, so page content moves across untouched — nothing is re-encoded or re-compressed, text layers and vector artwork survive intact, and scanned images keep their original quality. The trade-off is that the result is a *new* document, and a few structures attached to the old ones do not travel with the pages; the notes below list exactly which. Every step of the parsing and writing happens inside your browser tab, and the files you pick are never sent to a server.

How to use it

  1. Drop your PDFs onto the dashed box at the top, or click it to open the file picker. It is labelled "Select PDF files (multiple allowed)" — pick several at once or keep adding in batches.
  2. Check the list below. Top to bottom is the page order of the merged file. Each row has "Move up" and "Move down" arrows that shift it one position at a time, plus a cross to "Remove" it. There is no drag-and-drop reordering.
  3. Once the list holds at least 2 files, a "Merge PDF" button appears beneath it. With a single file the button is not rendered at all rather than shown greyed out — so if you cannot find it, you are one file short.
  4. Press "Merge PDF". The button switches to "Processing..." and is disabled while it works, which can take a few seconds on large inputs.
  5. When it finishes you get "Merged successfully — N pages" and a "Download" link; the saved file is always named merged.pdf. That page count is read back out of the finished document, so you can check it against the sum of your sources.
  6. Any later change to the list — adding, reordering, removing — makes the previous download link disappear immediately. That is deliberate: it stops you from downloading a file that no longer matches the order on screen. Press merge again after editing.

When it comes in handy

Bundle a tender or grant application into one attachment

The main submission, appendix A, the annexes and the financial statements each arrive as a separate PDF, and the portal accepts exactly one upload. Add them in the order the guidance specifies, arrange with the arrows, and export once. Because pages are copied verbatim, the signature stamps, scan watermarks and page dimensions the reviewer expects are all still there.

Reassemble a document your scanner chopped up

Sheet-fed scanners often flush a batch when the feeder empties, leaving scan_001.pdf, scan_002.pdf and so on. Adding them in filename order stitches the document back together. Note that this tool appends whole documents; it cannot interleave pages from two files, so it will not rescue a duplex scan that came out as one odd-page file and one even-page file.

Roll monthly reports up into a quarterly or annual pack

Twelve identically formatted monthly reports become one yearly compilation for the board or the shared drive. Merging does not normalise layout, so the two months that switched to landscape chart pages stay landscape rather than being squeezed into portrait.

Put a signed page back into the original contract

The usual workflow: print the contract, sign only the last page, scan that page on its own. Use the split tool to pull pages 1 to n-1 out of the original, then merge that with the scanned signature page for a complete file. Be aware this is only a content splice — if the original carried a digital signature, merging invalidates it without exception.

Limits and behavior

  • The three size limits are checked at different moments. The 50 MB per-file cap applies the instant you add a file: an oversized one is rejected on its own while the rest still join the list. The 100 MB combined cap and the 500-page cap are only verified after you press merge, so it is entirely possible for everything to look fine while adding and for "Total size exceeds the 100MB limit" to appear only on the click.
  • Because the output is a freshly created document, bookmarks and outlines, the title / author / subject fields in the document information, and any fillable form fields are all absent from the merged file. On top of that /Producer and /Creator are rewritten to pdf-lib, and /ModDate and /CreationDate are set to the moment you clicked. If you only need to turn some pages and want those structures to survive, reach for Rotate PDF instead, which edits the original in place.
  • Per-page geometry does come across: page size and the /Rotate flag are preserved page by page. A mix of A4 and A3, or portrait and landscape, stays exactly as it was — nothing is normalised to a single paper size and nothing is scaled.
  • The merged file can end up larger than the sum of its inputs. pdf-lib does not deduplicate resources shared between documents, so five reports that each embed the same 12 MB Chinese font produce a result carrying five copies of it. This tool also performs no compression or image downsampling whatsoever; shrinking the result is a separate job.
  • A file counts as a PDF only if the browser reports its MIME type as application/pdf. Files dragged straight out of certain network drives or archive utilities arrive with an empty type field, and those are refused with "Please upload a PDF file" even though the name ends in .pdf. Saving to local disk first and re-selecting normally clears it.
  • Password-protected PDFs are stopped at load time, including the owner-password kind that opens without any prompt and merely restricts printing and editing. There is no password field here, and no attempt is made to work around the protection.
  • The same file can be added twice — the list does not deduplicate. That is useful if you genuinely want a cover or disclaimer page to appear more than once, but it also means picking the same appendix twice by accident passes silently. Glance over the list before you merge.
  • What it does not do: interleave pages from two files, insert a single page at a chosen position inside another document (split it into two parts first, then merge three items), add page numbers, generate a table of contents, apply watermarks, or reduce file size.

Frequently asked questions

Why is the merged file bigger than all the originals added together?

Mostly because resources are not deduplicated. A PDF stores font subsets, images and colour profiles as internal objects, and copying a page brings its objects along, so two documents using an identical font end up with one copy each. It is most visible with CJK reports out of a layout application, where an embedded font subset alone can be several megabytes. A smaller factor is that the whole document is re-serialised on save, so a source that used a more compact cross-reference structure may gain or lose a little in the rewrite.

My bookmarks and fillable form fields vanished after merging. Can I get them back?

Not within this tool. An outline hangs off the document catalog and form fields are registered in a document-level AcroForm structure; neither belongs to a page, so a page-by-page copy leaves them behind. In practice you either rebuild the bookmarks afterwards in a reader that can edit them, or — if it is the fillability that matters — skip merging and send the blank form as its own attachment.

I added one file and there is no merge button anywhere.

With fewer than two files the button is not rendered at all rather than disabled, which is why there is no greyed-out control to find. Add a second PDF and it appears directly under the list. Incidentally, the message "Please select at least 2 PDF files" exists in the code but you can never actually trigger it on screen, precisely because the button is hidden until then.

How is the 100 MB total measured, and why was I not warned while adding files?

It measures the combined byte size of the source files at merge time, not the size of the output. Adding a file only compares it against the 50 MB per-file limit; the total is checked after you click, once each file has been read into memory, which is why the warning arrives late. If you hit it, merge in two batches and then merge the two intermediate results — that works as long as each intermediate stays under 50 MB.

Is the 500-page limit per file or across all of them?

Cumulative. Pages are counted as each source is read in, and the moment the running total passes 500 the operation aborts with "Page count exceeds the limit of 500" — no partial file is produced. So three 200-page documents (600 pages combined) fail while reading the third, even though each one is comfortably under the cap on its own.

Can I drag rows to reorder? I have twenty files to arrange.

Only the up and down arrows exist, moving one position per click, which is genuinely tedious at twenty files. The shortcut is to get the order right before you add anything: rename the sources with 01_, 02_ prefixes in your file manager, sort by name, then select them all and drop them in one go. The browser hands them over in the order you saw, which is usually already correct.

Is the merged PDF still searchable?

Yes, as long as the sources had a text layer. Merging relocates page objects, so the text, its font mappings and its coordinates are all preserved; searching, selecting and copying behave exactly as they did in the originals. The one exception is a source that was pure scanned imagery to begin with — it had no text layer, merging cannot invent one, and what that needs is OCR, which this site does not offer.