Merge PDF

This guide covers what this tool does, control by control: how it orders files, what it keeps, what it rebuilds and what it drops. For the general explanation of what a merge does inside a PDF, and the habits that keep merged files tidy, read how to merge PDF files free online without uploading them.

How to merge PDFs with this tool

Setting the order and choosing pages

The cards are the output order, left to right and top to bottom, and the number in the corner of each card is its place. Files are added in the order your browser hands them over, which is not always the order you picked them in, so check those numbers before merging.

Sort by name compares the numbers inside names as numbers. Plain text sorting, which many file managers and upload dialogs use, compares one character at a time, so page10.pdf lands before page2.pdf. Sort by name puts them the way you meant; in the checker below, the right-hand column is the order it produces.

Each card's page count is also a page picker. Type 1-3, 7 for pages 1 to 3 and 7, or 10- for page 10 to the end. Pages come out in the order you type them, so 5, 1-4 moves page 5 to the front, and a page typed twice is used once. A range with letters in it, or one that matches no page, stops the merge with a message naming the file.

What this tool does to bookmarks, links and annotations

With both bookmark options on, the outline of the result has one entry per file, named after the file without its extension, and each file's own bookmarks sit under it at their new page numbers. Turn off the first option and the files' bookmarks sit side by side at the top level; turn off the second and only the file names remain. If you took a page range, a bookmark whose page you left out is dropped, unless bookmarks under it survive, in which case it stays as a heading that goes nowhere.

A link to a website is stored on the page, and it comes across with the page. A link that jumps to another page in the same file points at that page's object, which copying would leave behind, so the tool rewrites each one to the page's new position, named destinations included. When a page range leaves the target page out, the link is removed rather than left pointing nowhere.

Forms, signatures and password-locked files

Form fields stay fillable. The tool gathers the fields on every page of the result into one form, taking its default font settings from the first file that has a form. Fields are addressed by name, so two copies of the same form, each with a field called full_name, can end up linked in some readers, where typing in one fills the other. If you need several completed forms in one PDF with every answer kept apart, flatten each one first. Flattening turns the typed values into ordinary page content.

A file with only an owner password, the kind that opens freely but restricts printing, copying or editing, merges normally. The tool loads a decrypting build of its PDF library from this site and reads the file in your browser. The result carries no password and none of those restrictions, so protect it again if they mattered.

A file that asks for a password to open cannot be merged here. Its card says password-protected, and Merge stops with a message that one of the files could not be read. Remove the password with the Unlock tool, using the password you have, then add the unlocked copy.

Images, page sizes and rotation

Each PDF page keeps its own size and rotation. Before copying, the tool writes each page's size, crop box, rotation and resources onto the page itself, because a page that inherits them from elsewhere in its source file would otherwise lose them in the copy. An A4 report followed by a Letter scan and a landscape spreadsheet makes a legitimate PDF with jumps in it, and every reader shows it that way.

Each image becomes one page. The tool turns a photo upright from its EXIF orientation, draws it to a canvas and embeds it: PNG as PNG, JPG and WebP as JPEG at quality 0.92, so a JPG is re-encoded once. The page takes the size of the first PDF page in the merge, turned to match the photo, so a picture sits in the document at the same size as the pages around it. With no PDF in the list it is Letter in the US and Canada and A4 elsewhere. The image is scaled down to fit and centred, and never enlarged beyond one pixel to a point, so a small graphic is not blown up into a blur. For margins or a different paper size, convert the photos with Images to PDF first, then merge its output.

• Sideways scans. Run the offending file through Rotate PDF first. Rotation is a per-page flag, so fixing it costs nothing in quality.
• Different paper sizes. Usually harmless on screen. For printing, the printer driver scales each page to the tray, which is normally what you want.
• Blank backs from a duplex scanner. Remove them in the Editor before merging rather than hunting for them afterwards in a longer file.

Large merges and error messages

There is no file or page cap in the code. The browser holds every source file, and the new document, in memory while it works, so the limit is your device's memory, and it is lower on a phone than on a laptop. If a very large merge stalls, merge in two halves and then merge the halves. Nothing is redrawn, so the pages are the same either way.

When something goes wrong, the tool says so in one of four ways:

• A card marked password-protected. The file needs its open password. See the section above.
• A card marked unreadable. The browser could not open it as a PDF or an image. Remove the card, then repair or re-export the file.
• “The page range for file name isn't valid. Use numbers and dashes, like 1-3, 7.” The range typed on that card has letters in it or matches no page.
• “One of these files could not be read. Make sure each is a valid PDF or image.” A file is damaged or needs an open password. The cards stay as they were, so remove the one at fault and merge again.

What stays on your device

The files are read with the browser's own file access and merged in the tab. The page fetches its scripts from this site, and for an owner-password file one more, the decrypting build of pdf-lib; none of those requests carries your files. To check, open the browser's developer tools on the Network tab before you merge and watch what is sent.

What to do after merging

The merged file is a normal PDF 1.7 document. It opens in Acrobat, Preview, every browser and every mobile reader, and it uploads to signing services and government portals without complaint. A few follow-up jobs are common enough to plan for:

• Too big for the portal or the inbox. Compress PDF shrinks the combined file in one pass, which gives a better result than compressing the pieces beforehand.
• The reader needs to know where they are. Add page numbers across the whole document so the numbering runs continuously through every source.
• It should not travel unprotected. Password protect the finished file, then share the password by a different channel.
• You merged too much. Split PDF pulls a page range back out without touching the rest.