Free PDF Compare Tool – Find Differences Between Two PDFs

How Comparing Two PDFs With a Word-Level Diff Works

A PDF compare tool reads the text of two documents and shows you exactly what changed between them, word by word, so you never have to read both files side by side hunting for a single edited clause. This is the sort of task that eats hours by hand and that people get wrong when tired. Two versions of a contract can differ by one deleted sentence; a policy document can look identical but have a changed number buried on page nine. The people who lean on this are lawyers checking a revised agreement, editors comparing manuscript drafts, developers reviewing two versions of a spec, and administrators confirming that a resent form matches the one they approved. Under the hood the tool uses pdfjs-dist to extract text and the diff library to compare it. For each document it opens the file with pdf.js, walks every page, and calls getTextContent, which returns the individual text runs that it joins into a per-page string. It then aligns the two documents page by page and runs a word-level diff on each pair, producing a sequence of parts marked unchanged, added, or removed. Additions are highlighted in green and deletions in red with a strike-through, shown inline so you read the new text with the old text crossed out in place. A per-page change count and a document total tell you at a glance how much moved, and a changes-only filter hides every page that is identical. Here is how a comparison plays out. Suppose page one of an original says the quick brown fox and the revised version says the slow brown fox. The tool extracts both strings, diffs them, and shows the unchanged words in normal text, quick struck through in red, and slow added in green, with the page reporting two changes, one removal and one addition. On a longer document, say a 10-page contract where only page six changed, the summary reports the total change count and, with changes-only on, jumps you straight to page six so you are not clicking through nine identical pages. An ignore-whitespace toggle collapses runs of spaces before diffing, so a document reflowed with different line breaks does not light up as changed when only the layout moved. The practical cases are exactly the high-stakes ones. A legal team compares the counterparty's returned contract against the version they sent, catching a quietly altered payment term before signing. A publisher's editor diffs a second draft against the first to see precisely which paragraphs an author rewrote. A compliance officer checks that a re-uploaded policy matches the approved wording, not just a similar-looking file. A product manager compares two releases of a requirements PDF to confirm which lines the client actually changed. In each of these, missing a single-word difference has real consequences, which is why a mechanical word-level diff beats a human skim. There is one limitation that matters most, and the tool handles it honestly. A scanned PDF is just images with no text layer, so text extraction returns nothing; rather than falsely reporting that the documents are identical, the tool detects the empty text and tells you the file is likely scanned and should be run through OCR first. Beyond that, this is a text diff, not a visual or layout diff, so it compares words rather than fonts, colours, or image positions, and a moved paragraph shows up as a deletion in one place and an addition in another rather than as a move. Ignore-whitespace is best left on for documents that reflow, and encrypted files need unlocking first. Because both text extraction and the diff run in your browser, two confidential contracts or draft manuscripts are compared entirely on your device and never uploaded to any server.

🔥 Trending Tools