Compare two PDFs and see exactly what changed
The Compare PDF tool lines up two versions of a document page by page and shows every difference: words that were added or removed, pages that were inserted, deleted or rewritten, and changes to pictures, charts, signatures or layout that text alone would never reveal. A colour overlay marks every changed pixel, so even a moved logo or a nudged decimal point stands out.
It is built for the moments when “I only changed a few things” needs checking: a contract returned by the other side, a revised quotation, a thesis after supervisor feedback, a new edition of a price list, or proofs from a printer. Summary numbers tell you at a glance how much changed, and a downloadable report lets you share the result.
Both PDFs are processed in your browser. Neither file is uploaded, so confidential agreements and unpublished work stay private.
How to compare two PDF files
- Drop the earlier version into “Original PDF” and the newer one into “Changed PDF”. Use “Swap files” if you put them the wrong way round.
- Tick “Ignore case” or “Ignore punctuation” if those differences do not matter, then press “Compare PDFs”.
- Read the summary and the page table: each row pairs an original page with its match in the changed PDF and labels it identical, text changed, visual change, replaced, inserted or removed.
- Click “View” on a row (or use the arrows to jump between changed pages). Switch between Overlay, Side by side and Slider, and adjust the pixel tolerance if anti-aliasing noise shows up.
- Read the highlighted text changes under the picture, then download the HTML or text report.
Compare PDF features
Smart page matching
Pages are paired by their content, not just their position, so an inserted page does not make every following page look different.
Word-level text diff
Each matched page is compared word by word with the Myers difference algorithm, with removed words struck through in red and added words in green.
Pixel-difference overlay
Both pages are rendered at the same scale and compared pixel by pixel. Ink only in the original turns red, ink only in the new version turns green, and changed areas are boxed.
Three viewing modes
Overlay for spotting changes, side by side for reading both versions, and a slider that wipes between them.
Shareable reports
Download a self-contained HTML report with the page table, highlighted text changes and difference images, or a plain-text report in [-removed-] {+added+} notation.
Private
pdf.js renders and reads both files on your device; nothing is sent to a server.
How the comparison works
First the text of every page is extracted. Pages are then matched with a dynamic-programming alignment that rewards pairs whose words overlap (measured with the Dice coefficient) and allows pages to be skipped on either side. Unmatched pages are reported as inserted or removed; when a removed page sits exactly where a new page appears, the pair is shown as “replaced”. Pages without any text, such as scans, are matched using small thumbnails instead.
Each matched pair is diffed word by word, and both pages are rendered and compared pixel by pixel. A pixel counts as changed when any colour channel differs by more than the tolerance, which filters out tiny anti-aliasing differences. Changed pixels are grouped into boxes so you can see at a glance where on the page to look.
What each status means
| Status | Meaning |
|---|---|
| Identical | Same words and no visible pixel differences |
| Text changed | Words were added, removed or edited — see the highlighted text |
| Visual change | The words are the same but something looks different: an image, a chart, colours, spacing or position |
| Replaced | An original page was swapped for a page with mostly different content |
| Inserted | The page exists only in the changed PDF |
| Removed | The page exists only in the original PDF |
Tips for accurate results
- Comparing scanned documents? There is no text to diff, so rely on the visual overlay, and raise the tolerance a little because two scans of the same page are never pixel-identical.
- If line breaks moved because a paragraph was re-flowed, the word diff still lines up the text correctly — only real word changes are highlighted.
- To compare plain text or code rather than PDFs, use the Diff Checker. To pull all the text out of one PDF, use PDF to Text.
- Before sending the final version, check it for hidden problems with the PDF Validator and tidy its properties with the PDF Metadata Editor.
Compare PDF FAQ
How do I compare two PDF files for differences?
Add the original PDF on the left and the changed PDF on the right, then press “Compare PDFs”. The tool pairs the pages, highlights changed words and shows a pixel overlay of every visual difference. You can download the result as an HTML or text report.
Can it detect pages that were added or deleted?
Yes. Pages are matched by content, so an inserted page is reported as inserted and the pages after it still line up with their originals. Deleted pages are reported as removed.
Does Compare PDF find changes in images and layout?
Yes. Every matched pair is rendered and compared pixel by pixel, so moved, recoloured or replaced images, logos, signatures and charts show up even when the text is identical.
Can I compare scanned PDFs?
Scanned pages have no text layer, so only the visual comparison applies. Pages are matched by thumbnail similarity, and you may need a higher pixel tolerance because scans differ slightly every time.
Are my PDFs uploaded to compare them?
No. Both files are opened and compared entirely in your browser and never leave your device.
Is there a page limit?
There is no fixed limit. Very long documents take longer, and pixel comparison in the summary covers the first 300 page pairs; text comparison covers every page.