Compare two PDFs
Load the old version and the new one of the same document and see at a glance the added lines in green, the removed ones in red and the changed words inside a reworded line. The files stay on your device.
.pdf files only
.pdf files only
🔒 Both PDFs are read on your device, nothing is uploaded anywhere.
How to use it
Put the version you already had on the left and the one you just received on the right, then press «Compare the two PDFs». Lines on a green background with a + sign exist only in the new version, lines on a red background with a minus sign only in the old one. When a line has been reworded you see it twice, one above the other, with only the changed words highlighted inside the line: that is the case where «within 30 days» quietly becomes «within 60 days». The «What to show» menu switches between the differences alone (with two lines of context around them) and the whole text, and at the end you can copy or download a .txt report to attach to an email.
Why the comparison covers the whole document instead of page against page
The text of each page is rebuilt into lines from the coordinates of every text chunk, the vertical one to tell where a line ends and the horizontal one to put the words back in order and to add a space where a gap calls for it. The lines of the two files are then aligned all together with an anchored diff, which starts from the lines that appear exactly once in both documents, the safest hooks available. Comparing page 3 with page 3 would have been easier and useless, because a single clause added at the top pushes everything else down and a document changed in one spot would come back as changed everywhere. The page shown is the one of the second file and, when the text has moved, the matching page of the first file is written next to it.
If the PDF is a scan the tool says so
A PDF can hold real text or just the photograph of a sheet of paper. In the second case there is nothing to read, and an honest comparison cannot answer «no differences» when it has seen nothing at all. If a file holds no text the page stops and says it, and if only some pages are scanned it lists them in a warning: those pages stay out of the comparison. Bringing them in needs an OCR first, that is optical character recognition.
What is not compared
The text and nothing else. Images, logos, graphic signatures, colours, fonts, margins and layout stay out, so two files with identical text can still look different. Tables and two-column texts are read line by line in the order the PDF stores them, and that order is sometimes not the one your eye follows: around those spots the comparison can report moves that are not real edits. For the same reason the page-number option is on by default, otherwise two extra pages would be enough to produce dozens of pointless differences. The limit is 400 pages per file.
The files never leave your device
Everything happens in the browser: no upload, no server, no copy left anywhere. That is the difference that matters when the two files are a contract, a quote or an expert report. You can even go offline before opening them and the page keeps working.