Compress a PDF to a smaller file size
Compressing a PDF reduces its file size; this tool does it by rendering each page and re-encoding it as a quality-controlled JPEG, entirely in your browser with no upload. The trade-off is that pages become images, so text is no longer selectable or searchable in the output. This method is most effective on scans and image-heavy documents, where savings of 60–90% are typical.
How it works
Most oversized PDFs are big for one reason: the images inside them carry far more pixels and far gentler compression than viewing requires. A phone "scan" often embeds a 12-megapixel JPEG per page — several megabytes to display what a reader views at roughly letter size on a screen. Fonts, text operators, and vector art are usually a rounding error by comparison.
This tool attacks that directly through rasterization. Each page is rendered with the pdf.js engine onto an off-screen canvas at a fixed target resolution — enough pixels for crisp on-screen reading and ordinary printing, but far fewer than a raw scan carries. The canvas bitmap is then encoded as a JPEG at your chosen quality factor. JPEG is lossy: it divides the image into 8×8 pixel blocks, transforms each with a discrete cosine transform, and discards the high-frequency detail your eye is least sensitive to. Lower quality factors discard more, which is where the size savings come from.
A new PDF is then assembled in which every page contains exactly one object: the JPEG of the old page, placed at the original page dimensions. This is why the result is smaller but also why it is flat. Text objects, embedded fonts, hyperlinks, form fields, and vector graphics no longer exist — they were baked into pixels during rendering. The output prints and displays like the original but behaves like a photo album of it: nothing is selectable, searchable, or clickable.
The corollary is worth stating plainly: for a PDF that is mostly live text, rasterization is counterproductive. A page of text stored as operators and a subsetted font might occupy 5 KB; the same page as a JPEG needs 60–150 KB. If your file is text-heavy and only mildly oversized, try the web optimize tool, which recompresses streams losslessly instead.
Worked example: a 28 MB scanned contract down to 2.1 MB
A 14-page contract scanned on a phone comes in at 28.4 MB — about 2 MB per page, because each page is a 4032×3024 JPEG saved at high quality. The upload portal it must go to caps attachments at 5 MB.
Running it through this tool at quality 0.7 renders each page at screen-oriented resolution and re-encodes it, producing a 14-page output of 2.1 MB — roughly 150 KB per page and a 93% reduction. Text remains perfectly readable on screen and in a laser print; faint shadows on the paper background actually diminish because JPEG smooths low-contrast noise. At quality 0.5 the same file lands at 1.4 MB, but faint ringing appears around small type, so 0.7 is the better choice when 2.1 MB already fits the limit. The original file is untouched on disk, which matters here: the compressed copy has no selectable text, so the original remains the authoritative version.
Compress PDF — frequently asked questions
Why is the text not selectable after compressing my PDF?
This tool compresses by converting each page into a JPEG photograph of itself, and a photograph contains no text objects. Keep the original file whenever you may later need to copy, search, or extract the text.
How much smaller will my PDF get after compression?
Scanned documents and image-heavy files typically shrink 60-90 percent because their oversized images are re-encoded at screen resolution. Lean text-only PDFs can get larger instead, since a picture of a text page takes more bytes than the text itself.
Can I compress a PDF without losing any quality at all?
Not with rasterization, which is inherently lossy. Lossless savings are only possible by restructuring the file, for example recompressing streams, which usually saves far less; our web optimize tool takes that approach.
Is my PDF uploaded to a server to be compressed?
No. The rendering and JPEG re-encoding run in JavaScript and canvas inside your browser, so the file never leaves your device. Large files are limited only by your machine's memory, not by an upload cap.
What JPEG quality setting should I choose when compressing?
Around 0.7 is a good default for scans and mixed documents; artifacts are hard to spot at reading size. Drop toward 0.5 only when hitting a strict size limit, and expect visible ringing around sharp text edges there.
Will compressing remove the fillable form fields from my PDF?
Yes. Because every page becomes a flat image, interactive elements such as form fields, links, and annotations are not carried into the output. Flatten or fill forms first if you need their contents to remain visible.