Compare PDF
Drop an old and a new version of a PDF and see every word that was added, removed, or changed — as a redline or side-by-side, with nothing ever leaving your browser.
Drop the older / original PDF (Version A)
How PDF version comparison works
Compare PDF is a browser-based tool that extracts the text from two PDF files — an older "Version A" and a newer "Version B" — and runs a word-level diff between them, so you can see exactly what changed without reading both documents side by side yourself. Added words are highlighted in green, removed words are struck through in red, and unchanged text stays neutral, giving you a quick, accurate answer to "what actually changed between these two drafts?"
You get two ways to read the result: a unified view, which shows one continuous document with insertions and deletions marked inline (the same format a code diff uses), and a side-by-side view, which splits Version A and Version B into two panes so you can read each one in context. A change counter with Next/Previous buttons jumps you straight to each edit, a page strip flags exactly which pages contain changes, and a similarity score gives you an at-a-glance sense of how close the two versions are before you dig into specifics.
Worth knowing up front: this is a text comparison, not a visual one. It reads the selectable text layer of each PDF and diffs the words — it won't catch a moved image, a swapped font, a recolored heading, or a reformatted table, since none of those change the underlying text. If a layout-level or pixel-level comparison is what you actually need, a desktop tool built for that (like Adobe Acrobat's Compare Files) is the better fit. For catching wording changes — which is what actually matters in a contract redline, a policy update, or a changelog review — text-level comparison is both faster and more precise, since it flags the exact words instead of a fuzzy "something changed here" box.
Everything runs locally using PDF.js inside your browser tab — both files are read and diffed on your own device, with no upload step and no server ever seeing their contents. That matters for anything sensitive: an unsigned contract, an HR letter, a draft press release, or internal pricing you don't want passing through a third party just to check what a colleague edited.
Common use cases
- Checking exactly what a counterparty changed in a returned contract or NDA
- Reviewing a collaborator's or editor's edits to a shared document
- Confirming a "final" PDF really matches the version you already approved
- Catching accidental text changes after re-exporting or re-printing a document to PDF
- Tracking wording changes between two revisions of a policy, spec, or terms-of-service page
Related tools
Compare PDF — frequently asked questions
Just the text. It pulls the selectable text out of both PDFs and runs a word-level diff — additions, removals, and unchanged text. It doesn't detect a moved image, a swapped font, or a reformatted table. For pixel-level visual comparison (catching layout or image changes), a tool like Adobe Acrobat's Compare Files is the right fit instead.
Version A should be the older or original file, Version B the newer one you're checking against it. Every addition and removal in the result is shown relative to that direction — swap them by mistake and green/red will read backwards, though the "Swap A ⇄ B" button on the results screen fixes that in one click without re-uploading.
Not directly — both files need a text layer for the diff to work. If either version is a scan or a flattened image-only PDF, run it through OCR PDF first to add searchable text, then compare the OCR'd output here.
By default, yes — case counts as a change, matching how most diff tools behave. If you only care about actual wording and want to ignore case differences, turn on "Ignore letter case" before comparing (or after, from the results screen — it re-runs instantly without re-uploading).
No hard limit. Everything runs on your own device, so the practical ceiling is your browser's memory and patience rather than a server quota — a few hundred pages compares in seconds, very large scanned-then-OCR'd documents take a bit longer.
Yes — the results screen has a "Download report" button that saves a plain-text file listing every addition and removal (marked with +/- like a code diff), plus a "Copy report" button if you just want to paste it into an email or ticket.
No. Both files are read and diffed entirely inside your browser using PDF.js — nothing is transmitted anywhere, which matters for contracts, HR letters, or anything else you'd rather not send through a third-party server.