Portals want 500 KB. Email wants under 25 MB. Type the number you need and the compressor searches resolution and quality until the file fits \Uffffffff on your own machine, with nothing uploaded.
Read and rewritten on your device. Nothing is sent anywhere.
The two methods work very differently. Pick the one that matches your file.
Leave blank to use the quality preset below instead.
With a target size, these are only the starting point \Uffffffff the search lowers them as needed.
Each attempt is shown, so you can see how it got there.
Advertisement
Almost every oversized PDF is oversized for one reason: it contains pictures at a far higher resolution than anyone will ever look at.
A phone camera photographs a page at 12 megapixels. Printed at A4 that is roughly 300 DPI, and on screen it is far more detail than the eye can use. Dropping to 120\Uffffffff150 DPI and re-encoding typically removes 70\Uffffffff95% of the file.
A PDF exported from Word is mostly font data and instructions. There is very little to squeeze, which is why a 200 KB text PDF stays near 200 KB whatever tool you use. Anything promising 90% off such a file is turning your text into a picture.
Rather than guess, the tool encodes a few sample pages at different settings, estimates the whole document, then builds and measures. If it overshoots it steps the settings down and rebuilds \Uffffffff usually landing within one or two attempts.
Every step \Uffffffff reading the PDF, redrawing the pages, writing the new file \Uffffffff happens inside this browser tab. There is no upload, no server copy and therefore no retention policy to trust. You can disconnect from the internet after the page loads and it still works.
Processing uses your own machine, so a 300-page scan takes longer than it would on a server, and a very large file can strain an older phone. That is the honest trade for not handing over the document.
Open the compressed file before you submit it. If small print has gone soft, raise the target or the starting resolution and run it again \Uffffffff a rejected application costs more than a slightly larger file.
No. The file is opened, redrawn and rewritten entirely inside your browser tab. Nothing is transmitted, so there is no copy anywhere to be deleted afterwards.
Type 500 in the target box with KB selected, and use the rebuild-as-images method. The tool tries a setting, measures the result, and steps down until the file is under your limit. It aims to land just under rather than exactly on the number, since portals reject anything over.
You probably have a text PDF and chose the structure-tidying method, which cannot do much with one. Text and font data are already compact. Switching to the rebuild-as-images method will shrink it a great deal, at the cost of the text no longer being selectable or searchable.
Only with the structure-tidying method. Rebuilding pages as images turns everything into a picture, so text can no longer be selected, searched or copied. For a scanned document that was never selectable anyway, you lose nothing.
There is a floor. Below roughly 60 DPI at the lowest quality, pages stop being readable, so the tool stops there and tells you the smallest size it could reach. If that is still too big, the usual answer is fewer pages rather than more compression.
The file never leaves your device, which is the strongest privacy position a web tool can offer. As with any web page, use a device you trust and check the compressed output before sending it on.
Every page has to be redrawn by your browser \Uffffffff there is no server doing it faster in the background. The tool renders each page once and re-uses that render while it searches for the right quality, so it will not redo work needlessly, but a few hundred pages at high resolution still takes real time on your own hardware. Closing other heavy tabs and choosing a lower starting resolution both speed it up.