Zero quality loss
Nothing is re-encoded: images, fonts and vectors are copied through as they are. The savings come from structure, not from degrading your pages, so output is pixel-identical.
Shrink PDF files by rewriting them with compressed object streams.
Loading tool…
The tool is loading its code on your device. This happens once and is cached for later visits.
Processed entirely on your device
Overview
Compress PDF files free online without touching page quality. Metadata is stripped and the file is rewritten with object streams — the size saving is shown up front.
Most PDFs are larger than they need to be, and almost none of the waste is in the pages you see. It is in the structure: uncompressed cross-reference tables, one object per line, duplicated metadata and producer baggage accumulated across every application that ever saved the file.
The document is parsed and rewritten with pdf-lib using compressed object streams — many small objects packed into a few deflated streams instead of thousands of plaintext lines — and the metadata dictionary is cleared. Pages, images and fonts are copied through byte-for-byte, which is why the visual result is identical and the operation is fast.
PDFs that were already written with object streams — most modern print-shop and LaTeX output — have little structural waste left. The honest answer for those is that they are already small, and the size delta shown per file says so plainly.
Clearing title, author, subject and keywords is part of the rewrite. If you need those fields preserved, edit them back afterwards; if you were compressing partly to scrub a document's history before sharing, this does that for free.
Step by step
Drop one or more PDFs onto the upload area, or click it to browse.
Press Compress PDF — each file is rewritten locally with compressed object streams.
Compare the before and after sizes in the result rows and the summary strip.
Download the compressed files; page content is untouched.
Why use it
What this tool is good for, and what it deliberately does not try to do.
Nothing is re-encoded: images, fonts and vectors are copied through as they are. The savings come from structure, not from degrading your pages, so output is pixel-identical.
Files that are already optimised shrink little, and the tool shows that instead of claiming a win. Documents bloated by metadata and uncompressed object streams shrink dramatically.
Title, subject, keywords and producer information are cleared on rewrite — useful when a document's hidden history should not travel with it.
Questions
Short, honest answers about quality, limits and privacy.