Pick a strength: balanced (hard to tell the difference) or strong (smallest file).
Wait for it to finish, then download.
This one needs an upload, and is deleted afterwards
Browsers cannot do this calculation, so the file goes to our server. It is deleted automatically 60 minutes after the job finishes, and the working copy lives in a memory filesystem that is wiped whenever the server restarts.
We do not log file names and do not log file contents — that is a hard rule in our engineering standards, not a setting.
What it does — and what it doesn't
What gets compressed is the images. Scans are images on every page, which is why they shrink most — measured on a 6-page, 6.8 MB scan: 30.5% on balanced, 17.6% on strong.
Text-only PDFs barely move. Those files are already small, and their bulk is fonts and vector objects. We do not touch that part — the only mature implementation carries a licence we cannot use, and we decided not to do it. A file that will not shrink is not a fault.
Compression is not reversible. Keep the original.
Strong mode can soften small print in scans. Try balanced first on anything that matters.
Common questions
Why did my file hardly shrink?
It is almost certainly a text PDF. Its size is in fonts and vector objects, not images. We do not subset fonts or simplify vectors, so there is nothing for us to squeeze. That is a deliberate trade-off.
Can I still search the text afterwards?
Yes. The text layer is untouched — only embedded images are downsampled and re-encoded.
