Compare PDFs

Upload the original as the first PDF and the revised version as the second. The tool pairs matching pages, then highlights added, deleted, modified, and moved content on each page pair, with counters showing how many changes of each kind were found.

Side-by-side shows the two pages next to each other with synchronized scrolling, which is easiest for reading text changes. Overlay stacks the revised page on top of the original with an opacity slider, so shifted or resized elements stand out, and the Flicker button switches the top page on and off to make small moves visible.

With OCR turned on, page images are run through Tesseract in your browser and the recognized text is compared, so scanned documents can be checked. OCR currently uses the English language model, and it's slower than comparing PDFs that already contain text.

Changes are grouped into text, image, header/footer, annotation, formatting, and background categories, each with its own toggle. Turn off the categories you don't care about, and use the text search box to find a specific changed word or phrase.

Pages are paired by content similarity rather than by page number, so an inserted page doesn't throw off everything after it. A page with no counterpart is reported as added or removed, and the page navigation steps through the pairs the tool built.

Annotation changes are their own category, so added or removed comments, highlights, and markup are flagged separately from body text and can be shown or hidden with the annotation filter.

The Export as PDF menu produces a PDF of the comparison. In side-by-side mode you can choose a split view, alternating pages, or only the left or right document with highlights; in overlay mode it exports the overlay view.

No. Rendering, page pairing, OCR, and change detection all run in your browser with WebAssembly and web workers, and nothing is transmitted, so contracts and other confidential drafts stay on your machine.