Upload multiple PDFs, put them in order, and combine them into one file. Optionally scale every page to one size, so a scanned page doesn't come out four times bigger than the rest, and compress the result. Nothing leaves your browser.
// Your merged PDF will appear here
Merging uses pdf-lib to copy every page from each uploaded file, in the order you set, into one new PDF — this step is always lossless regardless of compression mode. If you pick a page size, pages that don't match it are scaled to fit and centred on that size, which is also lossless outside Aggressive mode. "Safe" compression re-saves that merged file with optimized internal structure (object streams, stripped metadata), which typically saves a modest amount with zero visual change. "Aggressive" compression goes further: each page is rendered to a canvas via PDF.js and re-encoded as a JPEG at the quality and resolution you choose, then rebuilt into a new PDF one image per page. All three modes run locally — you can confirm that by checking the Network tab in dev tools while a batch runs.
Each file is capped at 50 MB and each batch at 20 files, enforced up front. Aggressive mode additionally caps out at 200 total pages across the batch, since rendering every page to a canvas is meaningfully more work than copying it — for bigger jobs, use "Safe" mode (page count isn't limited there) or split into multiple aggressive-mode runs.
Text and vector content in a PDF is already compact — most of a PDF's size usually comes from embedded images and fonts, and "Safe" mode doesn't touch either of those; it only cleans up document structure and metadata. If your PDFs are text-heavy (reports, invoices, contracts), that's genuinely most of what a lossless pass can do. If they're scanned pages or full of photos, "Aggressive" mode will do far more.
When the PDF is mostly images to begin with — scanned pages, photo-heavy brochures, exported slide decks — and you specifically need a smaller file over a searchable one. It won't help (and will actively hurt) a text-heavy document where you or a colleague need to select, search, or copy the text afterward.
150 DPI at 0.7 quality is a reasonable default — legible on screen and in most printouts. Drop to 96 DPI / 0.5 quality if you just need the smallest possible file for an email attachment and don't care about print quality. Go up to 220 DPI / 0.85 if the pages contain small text you need to stay readable after compression.
No — pages inside each uploaded PDF keep their original order. The file-level order you set (via the arrows) only controls which whole file's pages come before which other file's pages in the final output.
No. Password-protected or encrypted PDFs generally can't be read without the password, so they'll fail to load here the same way they'd fail to load in most PDF tools — decrypt the file first if you need to include it.
This can happen in "None" or "Safe" mode when the source files share a similar structure — pdf-lib doesn't deduplicate identical embedded fonts or images across separate source documents, so a font embedded in three input files gets carried over three times. If the size difference matters, "Aggressive" mode sidesteps this entirely since it rebuilds every page from scratch as a single image.
By default every page is copied at exactly the physical size it had in its source file, so a page that looks huge or tiny was already that size before merging, most often because it started as an image converted to PDF at an assumed 72 pixels per inch. The result panel tells you when the merged file has mixed sizes and which pages are the odd ones. To fix it, set Page size to Match the most common size, Match the first page, A4 or US Letter and merge again. Pages that don't match are scaled to fit that size and centred, keeping their proportions; landscape pages stay landscape. This post explains where the odd sizes come from.
No. In None and Safe mode the page content is scaled with a transformation, not re-rendered, so text stays sharp, selectable and searchable, links and form fields move with the page, and the file stays about the same size. That's the difference from the usual workaround of printing the merged file to a PDF printer with a fit-to-page option, which rebuilds the whole file and can double its size or flatten forms. In Aggressive mode the pages are images anyway, and each one is placed on the chosen page size.
In a PDF, a form field is identified by its name, and a field with one name has one value. Most combine tools join same-named fields, so ten copies of one form end up as one field showing one answer everywhere. This tool keeps fillable fields but renames them per file (for example w9-alice_name and w9-bob_name), so each copy keeps its own answers, or flattens them if you choose. For a batch of filled forms, the PDF form merger also lists the field names your files share before you merge. Why combining filled forms changes the answers explains the mechanism.
Yes — upload just the one file. With only one file in the list there's nothing to merge, so the tool simply applies whichever compression mode you picked to that file on its own.