Skip to content
OmniConvert

Compress PDF

Make a PDF smaller by re-encoding the images inside it. Text stays text, and the whole thing happens in your browser — the file is never uploaded.

Runs on your deviceFree, unlimited

Drop files here

or

Accepts .pdf · One file at a time · up to 150 MB each "Upload" and "download" here mean your browser reading a file from this device and writing one back to it. Nothing is sent to a server — open your browser's network panel and watch: no request carries your file.

FAQ

Will my scan come out blurry?

That depends on the level, and the honest lever is resolution rather than quality. A page photographed by a phone is often 4000 pixels on its long edge, which is around 340 DPI for A4 — several times what a screen shows and more than a photocopier resolves. The middle setting brings that to 2000 pixels, about 170 DPI, which still looks clean on paper; the light setting keeps 2600 and the strong one drops to 1400, which is where text in a scan starts to soften. Nothing else on the page is touched at any level. For scale: a nine-page file whose images were 3400x4400 went from 6.67 MB to 0.73 MB on the middle setting, and looked identical on screen — though most of that came from the downscale, so a scan that is already near 2000 pixels will save far less.

Why did my file barely get smaller?

Almost certainly because it has no pictures in it. A text document is mostly font data and drawing instructions, and neither of those can be re-encoded — a PDF is large because of its images, and if there are none there is nothing here to take out. Rather than hand you back a same-sized file with a success message, the result says how many images were actually re-encoded, including when the answer is none.

Does the text get turned into a picture?

No, and this is the difference worth understanding. Some compressors get their impressive numbers by rendering each page to one flat image — the file shrinks a lot and the document stops being a document: text cannot be selected, searched or read by a screen reader, and our own PDF-to-text tool would find nothing in it. This one only replaces the image streams, so text stays text and vector drawings stay sharp at any zoom.

Are there images it cannot compress?

Yes, and they are counted rather than quietly skipped. We re-encode the images the browser can decode by itself, which in practice means the JPEGs that make up the overwhelming majority of scanned pages. JPEG 2000, the bilevel fax and JBIG2 codecs used by some scanners, and raw uncompressed samples all need a decoder we would have to ship or a colour space we would have to reinterpret, and getting either wrong silently ruins the picture. Those streams are left exactly as they were.

Do bookmarks and form fields survive?

They do. This edits the document you gave us and saves that same document back, so the outline, the internal links and any fillable form stay intact — the same as rotating and stamping. The tools that rebuild a document from copied pages, merging and splitting, keep the bookmarks but not the internal links.

Can I compress to an exact file size?

No, and we would rather say so than pretend. Hitting a target like "under 2 MB" means compressing, measuring, adjusting and trying again, several times, on a file that may not be able to reach the target at any quality. The three levels are predictable instead: each states what it does to the images, and you can see the result before you download it. If the middle setting is not enough, the strong one usually is.

PDF converter

PDF tools

PDF converterAbout PDF