How can a PDF be compressed without uploading it?

The tool re-renders each page on your device using Mozilla’s pdf.js, re-encodes it as an optimised JPEG, and rebuilds the PDF with pdf-lib — all in browser JavaScript. Your file never touches a server.

Why did my text become non-selectable after compression?

Browser-side compression rasterises pages — each page becomes an optimised image. That’s what makes scanned PDFs so much smaller, but it means text can no longer be selected or searched. If you need selectable text, keep the original for editing and use the compressed copy for sharing.

Which PDFs compress well — and which don’t?

Scanned documents, photo-heavy reports and image-based PDFs often shrink 60–90%. Text-only PDFs are already highly efficient and can even get bigger after rasterisation — the tool detects this and tells you to keep the original instead.

What compression level should I choose?

Balanced is the best default — big savings with quality that’s fine for reading and printing. Use Light when the document must stay crisp (contracts, forms you’ll re-scan). Use Strong only for chat/email sharing where a slightly soft look is acceptable.

Can I compress a PDF to a specific size, like 200KB?

Not to an exact target — compression results depend on page content. Try Balanced first, check the result size, then step to Strong if you need it smaller. For exam-portal photo/signature limits, our Resize Image to 20KB tool targets exact sizes for images.

Is there a file size limit?

No artificial limit. Since everything runs on your device, very large files just take longer, page by page, with a progress indicator. Memory on a typical laptop handles hundred-page scans fine.

Is this really private?

Yes. Open the browser’s network tab while compressing — you’ll see no upload requests. The PDF is read, compressed and rebuilt entirely in local JavaScript, which is also why there’s no queue and no “file deleted after X hours” policy to trust.

PDF Compressor

Shrink scanned and image-heavy PDFs in your browser — honest results, no uploads.

Files never leave your device Instant — no upload waitFree, no sign-up

Heads up: browser-side compression re-renders pages as images. Scanned and image-heavy PDFs shrink dramatically, but text becomes non-selectable. Text-only PDFs are usually small already — compressing them here can even increase size.

Choose a PDF or drop it hereCompressed on your device — never uploaded

or import from

Compression level

Good quality · big savings

How this PDF compressor works — honestly

Most “compress PDF” sites upload your document to a server running heavyweight tools, then hand back the result. This one is different: it compresses entirely in your browser, soyour PDF never leaves your device. To achieve that, it re-renders each page with Mozilla's pdf.js and rebuilds the document with optimised JPEG pages. The trade-off — and we say this up-front rather than in fine print — is that pages become images: text is no longer selectable. For scanned and image-heavy PDFs (the files people actually need to shrink), that trade-off costs nothing and savings of 60–90% are typical.

When to use it

Scanned documents from a phone scanner app or office copier · photo-heavy reports and portfolios · signed agreements that were printed and re-scanned · any PDF that's over an email or portal upload limit. Before/after sizes are shown instantly, so you can judge the result before downloading. If a text-only PDF would get bigger, the tool detects it and tells you to keep your original — no fake “compressed!” badge on a larger file.

Choosing a compression level

Light renders at high resolution with gentle JPEG compression — crispest result, modest savings. Balanced is the recommended default: a 1.5× render with 70% JPEG quality that reads cleanly on screen and in print. Strong pushes for the smallest possible file for chat apps and strict upload limits, with visible softening on fine text. Switching levels re-compresses immediately so you can compare all three in seconds.

Related tools

Need to shrink images instead? Use the Image Compressor. Splitting out only the pages you need with PDF Split before compressing is often the single biggest size win — a 3-page extract of a 60-page scan is 95% smaller before compression even starts. And JPG to PDF rebuilds a PDF from images once you've optimised them.

Frequently Asked Questions

How can a PDF be compressed without uploading it?
The tool re-renders each page on your device using Mozilla’s pdf.js, re-encodes it as an optimised JPEG, and rebuilds the PDF with pdf-lib — all in browser JavaScript. Your file never touches a server.
Why did my text become non-selectable after compression?
Browser-side compression rasterises pages — each page becomes an optimised image. That’s what makes scanned PDFs so much smaller, but it means text can no longer be selected or searched. If you need selectable text, keep the original for editing and use the compressed copy for sharing.
Which PDFs compress well — and which don’t?
Scanned documents, photo-heavy reports and image-based PDFs often shrink 60–90%. Text-only PDFs are already highly efficient and can even get bigger after rasterisation — the tool detects this and tells you to keep the original instead.
What compression level should I choose?
Balanced is the best default — big savings with quality that’s fine for reading and printing. Use Light when the document must stay crisp (contracts, forms you’ll re-scan). Use Strong only for chat/email sharing where a slightly soft look is acceptable.
Can I compress a PDF to a specific size, like 200KB?
Not to an exact target — compression results depend on page content. Try Balanced first, check the result size, then step to Strong if you need it smaller. For exam-portal photo/signature limits, our Resize Image to 20KB tool targets exact sizes for images.
Is there a file size limit?
No artificial limit. Since everything runs on your device, very large files just take longer, page by page, with a progress indicator. Memory on a typical laptop handles hundred-page scans fine.
Is this really private?
Yes. Open the browser’s network tab while compressing — you’ll see no upload requests. The PDF is read, compressed and rebuilt entirely in local JavaScript, which is also why there’s no queue and no “file deleted after X hours” policy to trust.