About Compare PDF
Compare PDFs finds every change between an old and a new version of a document with a word-level text diff. It extracts the text from both files, aligns them, and reports each addition, removal and edit — with character-level highlighting inside edited passages — then lets you download a comparison report as a PDF. It all happens in your browser; neither file is uploaded.
- No uploads
- Browser-only
- Works offline
- 100% free
How it works
- 1
Load both versions
Drop the old version on the left and the new version on the right, so additions and removals are labelled from the right direction.
- 2
Run the semantic diff
Click ‘Compare PDFs’. Both files are read, their text is extracted page by page, and a word-level diff is computed locally.
- 3
Review and export
Scan the change list — filter it with the search box — then download a PDF comparison report if you need a record.
What a text diff catches — and what it can't see
Compare PDFs is a semantic, word-level diff of the extracted text. That makes it excellent for catching real content edits between contract drafts or report revisions, and useless for changes that never touch the text. Knowing which is which saves you from trusting a ‘no differences found’ that only means ‘no text differences found’.
| Type of change | Detected? | Why |
|---|---|---|
| Added, removed or edited wording | Yes | The word-level diff aligns and compares the text content |
| A number, date or name changed | Yes | Character-level highlighting pinpoints the edited characters |
| Font, colour or spacing changed | No | The text is identical, so the diff sees no change |
| A replaced image or logo | No | Images carry no extractable text to compare |
| Two scanned (image-only) PDFs | No | There is no text layer to extract — OCR them first |
A pixel-overlay ‘Visual’ mode that would catch layout and image changes is marked Coming soon in the tool. For today, treat a clean result as ‘text is identical’, not ‘files are identical’.
Compare PDFs versus a plain text diff
If your two versions are already plain text, the general-purpose text diff is faster and shows a classic side-by-side. Compare PDFs earns its place when the content lives inside PDFs: it does the text extraction for you, aligns wording that has reflowed across pages, and produces a downloadable PDF report you can attach to an approval. Use the text diff for code, config and copy-paste snippets; use Compare PDFs when the source of truth is the document itself and you need an auditable record of what changed. When both files are scans, run OCR PDF first so there is a text layer to compare.
Frequently asked questions about Compare PDF
Does it compare visual layout or just the text?
Right now it compares the text content — a semantic, word-level diff that catches added, removed and edited wording anywhere in the document. A pixel-overlay ‘Visual’ mode is marked Coming soon in the tool; for now, changes to fonts, spacing or images that don't alter the text won't show up.
Why does it say my PDFs are scanned and can't be compared?
The diff needs selectable text. If both files are image-only scans with no text layer, there's nothing to extract and align. Run each one through OCR PDF first to add a searchable text layer, then compare the results.
What's in the downloadable comparison report?
A PDF listing every change — edits, additions and removals — with the old and new text for each, plus the file names you compared. It's built on demand when you click download, so you get a shareable audit trail of exactly what changed between the two versions.
Will it match up text that just moved to a different page?
The diff aligns text sequences rather than comparing page-for-page, so a paragraph that shifted position is usually recognised as unchanged rather than flagged as a delete-plus-add. Heavily reflowed documents can still produce noisier results than lightly edited ones.
Privacy, offline use, browser support, and pricing questions are answered on the site-wide FAQ.