Drop a PDF, pick a compression preset and download a smaller file. The embedded images are re-encoded locally in your browser — nothing is ever uploaded.
Files never leave your device
Good quality at a much smaller size. Very large images are scaled down to 1600 px.
Drop PDFs to compress, or click to browse
PDF
What This Tool Does
This tool makes a PDF file smaller by recompressing the images embedded inside it, without ever sending the document anywhere. Drop one or more PDFs onto the page, choose a compression level, and download each result with its exact savings shown.
Most oversized PDFs are oversized because of images: photos inserted at full camera resolution, scans stored with generous quality settings, or screenshots saved as uncompressed bitmaps. The compressor re-encodes those images as efficient JPEGs at your chosen quality, and — on the Balanced and Strong presets — scales down images that are far larger than any screen or printer needs. Text, fonts, vector drawings, and page layout are copied through untouched.
Tip:Start with Balanced. If the file needs to be smaller still, re-run the original through Strong — it caps image resolution at 1024 px, which is fine for reading on screen. Use Light for documents headed to a printer.
Why This Runs Entirely in Your Browser
Most online PDF compressors upload your document to a server, process it there, and let you download the result — meaning a copy of your file briefly lives on someone else's machine. This tool doesn't. The compression engine is a WebAssembly module that runs inside your browser tab, so your PDF's bytes never cross the network. For contracts, invoices, ID scans, and medical documents, that difference matters.
It also means there's no file-size cap imposed by a server, no queue, and no account. The only limit is your browser tab's memory, and a warning appears when a file is large enough for that to matter.
What Gets Compressed — and What Doesn't
The tool is deliberately conservative. It re-encodes ordinary photographic images (JPEG and raw bitmap data) where quality-for-size trading is safe and predictable. Image types that can't be rebuilt without risk — JPEG 2000, fax-encoded (CCITT) scans, JBIG2, CMYK press images, stencil masks — are passed through byte-for-byte identical rather than gambled with. If nothing inside a PDF can be safely recompressed, you get back a file no larger than the original.
That conservatism is why the output always opens correctly: nothing in the document is touched unless the engine fully understands it.
Picking a Compression Level
Light re-encodes images at high JPEG quality with no downscaling — the result is visually indistinguishable in almost all cases and still meaningfully smaller. Balanced, the default, drops quality to a level that looks clean on screen and in normal printing, and scales images down to at most 1600 px on their longest side. Strong pushes both dials further — noticeably smaller files, image resolution capped at 1024 px — and is the right choice when an attachment limit is the hard constraint.
Frequently asked questions
How much smaller will my PDF get?
It depends on what's inside. PDFs dominated by photos or scans — which is most large PDFs — routinely shrink by 40 to 90 percent, because the images inside are re-encoded at a more efficient quality and optionally scaled down. A PDF that's mostly text and vector graphics has little image data to recompress, so it may barely change. The result card shows the exact before and after size for every file.
Will compressing reduce the quality of my document?
Text, vector graphics, and page layout are untouched — compression only re-encodes the raster images embedded in the document. On the Light preset the difference is rarely visible. Balanced and Strong trade progressively more image sharpness for smaller files, and Strong also scales very large images down, which you might notice if you print the document at high resolution.
Is anything uploaded to a server?
No. The compressor runs as a WebAssembly module inside your browser tab. Your PDF is read, recompressed, and handed back as a download entirely on your device — its bytes never travel over the network. That makes this safe for contracts, statements, medical records, and anything else you'd rather not send to a stranger's server.
Why did my PDF barely shrink?
The most common reason is that there's simply little to compress: the file is mostly text, or its images are already stored at a low quality. The tool also deliberately skips image types it can't rebuild safely — for example JPEG 2000, fax-encoded scans, and CMYK print images — rather than risk corrupting them. Whatever it skips is left byte-for-byte identical, and the output is never larger than the input.
Which compression level should I pick?
Balanced is the right default: images are re-encoded at a quality that looks good on screen and in normal printing, and only unusually large images are scaled down. Pick Light when the document must print at full quality. Pick Strong when the file has to squeeze under an email attachment limit and on-screen reading is what matters.
Can I compress a password-protected PDF?
Not yet. Encrypted PDFs are detected and rejected with a clear error rather than producing a broken file. Remove the password protection first with another tool, then compress the unlocked copy.
Can I compress several PDFs at once?
Yes — drop as many files as you like and each is compressed independently with its own result card and download. Files are processed one after another in a background worker, so the page stays responsive while a large batch runs.