Compress a PDF
Recompress the photographs inside the document.Nothing is uploaded, so there is no wait.
or drag them here
One at a time · Files stay on your device
What exactly gets smaller?
Only the photographs the document carries. In a scan, or a deck built from screenshots of photos, the images are very nearly the whole file, so that is where the size goes. Text and fonts are left alone — subsetting them wrong is how a PDF ends up unreadable.
Sometimes nothing changes. Why?
A document made of text has no photographs to recompress. Losslessly stored graphics such as screenshots are skipped on purpose too, because baking them into JPEG smears the lettering they are made of. The result says so when that happens, so you can tell it apart from a failure.
Why cap the resolution?
An A4 page scanned at 600 dpi is around 5000 px on its long edge. Neither a screen nor an office printer ever asks for that. Capping at 2000 px usually takes a large bite out of the size with nothing visible to show for it.
Can I run it more than once?
Yes. Every run starts again from the original bytes, so trying a few quality settings does not stack losses. Your original file is untouched, and the result is saved beside it with -small on the end.
Where do the files go?
Nowhere. Your browser does the work, and this page never sends files to a server. Try it with your internet disconnected and see it still work.