Make a PDF small enough to email or to get past an upload limit. It's free, and the file is compressed inside your browser, never on a server.
Drop a PDF here or click to choose one
Scans and PDFs full of photos compress best
Combine files, delete extra pages, or split a document.
Almost all the weight in a large PDF is images. Scanned contracts, photographed IDs and slide decks full of screenshots lose most of their size, because the compressor re-encodes every picture inside them at a lower resolution and a lower JPEG quality. A report made of text and vector charts was tightly packed when it was written, so a compress pass often saves a few percent on it, sometimes nothing, because there was little to remove in the first place. The three levels set how hard the image pass pushes: Light keeps print resolution, Medium aims at reading on screen and Strong is for getting under a hard upload limit.
Visa applications, tenders and job boards often cap uploads at 100 KB, 200 KB or 2 MB. There is no target-size field here, so start on Strong and read the resulting size before you download. If the file is still too heavy the problem is usually page count, since a twenty-page scan will not fit in 200 KB at any setting, so cut it down with delete PDF pages or split PDF and compress what is left. When the source is a pile of photos and not yet a PDF, shrink the images first with our image compressor and assemble them afterwards, which gives you more control than compressing the finished document.
Most free PDF compressors upload your document to a server, work on it there and promise to delete it later. Here the file is read and rewritten inside the tab by a WebAssembly engine that loads the first time you press Compress, which costs a few seconds once and nothing after that. When the PDF is a signed contract, a bank statement or a passport scan, keeping it off someone else's server is reason enough on its own. There is no page limit, daily quota or account, and the only ceiling is your own machine: a very large document can take a couple of minutes, and the run gives up after three.
Drop the file in, pick Light, Medium or Strong and click Compress PDF. The page shows the old and new size side by side with the percentage saved, and you download the smaller file from there without an account and without uploading anything.
That depends almost entirely on what is inside it. Scans and photo-heavy documents lose most of their weight because their images are re-encoded at a lower resolution and quality, while a PDF of mostly text and vector drawings is compact already and saves little.
Yes. The compressor rewrites the file structurally instead of turning each page into a picture, so text stays selectable and searchable and links keep working. If that engine fails to load, a fallback pass rasterizes the pages, and then the text layer is gone.
You can try, but there is no target-size box, so you aim by hand. Run it on Strong, and if the file is still over the limit, remove the pages you don't need or split the document and compress again, since page count is the usual culprit.
The tool never hands back a file bigger than the one you gave it, and a text-only or already-optimized PDF often gains nothing at Light or Medium, so move up a level. If Strong gains nothing either, there is no slack left in the document.
No. Compression runs in a WebAssembly worker inside the browser tab and the document stays on your device. The first time you press Compress, expect a wait of a few seconds while that engine loads.