PDF Compare

About PDF Compare

This tool compares two PDF files and shows their differences. It compares file properties, metadata, page count, and file size. All processing happens in your browser for complete privacy.

What does the tool compare?

The tool compares file name, file size, number of pages, PDF version, and metadata (title, author, etc.).

Can it compare content?

This tool compares file-level properties and metadata. For detailed content comparison, you would need specialized diff tools.

What is the file size limit?

Each PDF file must be under 50MB.

Is my data safe?

Yes! All comparison happens in your browser. Your files are never uploaded to any server.

Can this tool detect text changes between two PDF versions?

Yes. The tool now does page-by-page extracted-text comparison: it pulls the text layer from each PDF, splits it into lines per page, and runs a line-level diff so you can see exactly which lines were added (+) or removed (-) on each page, with totals for changed pages and added/removed lines. This is a textual (content) diff, not a visual or OCR diff — it compares the words, not the rendered pixels or layout, which is what you usually want for contract, spec and report revisions.

Does it work on scanned or image-only PDFs?

Only if there is a real, selectable text layer. A scanned document that is just an image of a page has no extractable text, so the text diff will report 'no text layer'. If the scan was run through OCR, results vary with OCR quality — recognition errors can show up as false added/removed lines. For reliable text comparison, use PDFs exported from Word, LaTeX, Chrome or similar that keep a true text layer.

Is the text comparison affected by fonts, layout or page formatting?

No. The diff works on the extracted text, not on positioning, so font changes, margins, reflow and most layout shifts do not create false differences as long as the same words remain in the same reading order. Two PDFs that look very different but contain identical wording will diff as unchanged; conversely, edited wording is flagged even if the layout is untouched. Heavy column or table re-ordering can change reading order and may surface as moved lines.

Why does the size differ even if the document looks identical?

PDF size depends on: image compression level (JPEG vs Flate), embedded fonts (subset vs full), object streams enabled, linearization (web-optimized), encryption, redaction marks, change tracking. Two visually identical PDFs can differ by 5x in size based on these settings. Use a tool like Smallpdf to compress before comparing.

What metadata fields are compared between the two files?

Title, Author, Subject, Keywords, Creator (the app that authored content), Producer (the library that wrote the PDF), CreationDate, ModDate, and PDF Version (1.4, 1.7, 2.0). Differences in Producer often reveal which library generated each PDF (LaTeX, Word, Chrome, Acrobat).

Is the maximum 50MB per file a hard limit?

It is a soft browser-memory limit to prevent tab crashes. PDF.js loads the entire file into RAM. Modern laptops handle up to 200MB; mobile phones struggle above 30MB. For huge PDFs, split with pdftk or use a desktop tool like PDF24 Creator or qpdf for comparison.