Papyx
Compress a PDF, knowing what it costs
Compressing resamples the pages as JPEG: three quality levels, plus a greyscale option that saves a great deal more on a scan. It works where the weight comes from images — which is to say on very nearly every scanned document.
It has to be said plainly, and the app says it too: the result has no selectable text left. On a born-digital document, where the weight is text and vectors, this is the wrong tool — the others do not touch the content at all.
- Reads
- Writes
- Engine
pdf.js- Protected documents
- accepted, with their password
Compress a PDF offline, with no size limit
-
Open the PDF
If it is protected, type its password on its card.
-
Pick a level
Three levels of compression, from cautious to aggressive.
-
Greyscale, if it helps
On a black-and-white scan, the option cuts the weight again.
-
Compare and save
The resulting size is shown before you save anything.
Worth knowing
-
Three levels
A readable setting rather than a JPEG quality slider to interpret.
-
Greyscale
An extra option, decisive on monochrome scans.
-
No size limit
A 500 MB PDF goes through like any other: nothing is uploaded, so there is no quota.
-
Protected documents accepted
The tool goes through the rendering engine, so it takes a password.
Limits
-
The result is a series of images: the text is no longer selectable, searchable or copyable.
-
On a born-digital PDF (text and vectors), compressing can make the file heavier rather than lighter.
-
There is no lossless compression in Papyx: that is a deliberate choice, not an omission.
Questions about “Compress”
Does compression keep the text selectable?
No. Compressing resamples each page as JPEG, so the document that comes out is a series of images. That is the trade for the weight saved, and the app says so before the job runs.
Is there a size limit?
None. Nothing is uploaded: the file is read from your disk and processed inside the app, so there is no quota, no queue and no account.
Which documents does it work best on?
Scans, and any PDF whose weight comes from images. On a born-digital document, where the weight is text, it is not the right tool.