Compress PDF without uploading
Reduce PDF file size right in your browser.
Drop a PDF here, or click to upload
Compression runs locally in your browser
This compresses a PDF without uploading it. Each page is rendered locally with pdf.js, re-encoded as an optimised JPEG at the quality level you choose, and reassembled into a new PDF with pdf-lib. Because the work happens in your browser, the document never leaves your device — and because the method is rasterisation, it is extremely effective on image-heavy and scanned files and comparatively weak on pure text. Expect roughly 60 to 90 percent off a scanned document, where each page was already a photograph, and very little off a word-processor export whose text is stored as compact font instructions. The trade-off is explicit rather than hidden: rasterising turns text into a picture of text, so the output is no longer selectable or searchable. Use the light preset on text-heavy files to limit that damage, and skip this tool entirely when the text layer must survive.
About the PDF Compressor
Compress a PDF without uploading it to anyone's server. Each page is rendered locally with pdf.js and re-encoded at the quality you pick, then reassembled into a smaller document — all inside your browser tab. Compare the original and compressed sizes before you decide whether to download.
How to use the PDF Compressor
- 01Drag and drop your PDF, or click to choose it from your device.
- 02Pick a compression level: light, balanced, or strong.
- 03Click “Compress PDF” and wait a few seconds while it processes locally.
- 04Compare the original and compressed sizes, then download the smaller file.
Why use our PDF Compressor
- 01
Real browser compression
Pages are rasterized and re-encoded as optimized JPEGs to cut file size dramatically.
- 02
Adjustable quality
Choose light, balanced, or strong to match the trade-off you need.
- 03
Before & after sizes
See the original size, new size, and saved percentage at a glance.
- 04
Private by design
Compression runs in your browser — your PDF never leaves your device.
What does compressing a PDF actually do to the file?
Most of a large PDF's weight is images, not text, so the biggest available saving comes from re-encoding those images at a lower resolution and quality. This tool takes the most direct route to that: it renders every page to a canvas at a controlled scale, exports the canvas as a JPEG, and builds a fresh PDF where each page is that single flattened image. The consequence is worth understanding clearly, because it is a genuine trade-off rather than a free win. A scanned document — which was already just an image per page — loses essentially nothing conceptually and can shrink by 60 to 90 percent. A text-heavy PDF generated from a word processor is different: its text is currently stored as compact font instructions, and converting it to a picture of text both increases the size of that component and makes the text non-selectable. On such documents the tool may save little, and on some it can make the file marginally larger.
Choosing between light, balanced and strong
The three presets change the rendering scale and the JPEG quality together. Light keeps a higher resolution and quality, which suits documents you still need to read closely or print — legal paperwork, technical drawings, anything with small type. Balanced is the sensible default for sharing a scan by email, cutting size substantially while keeping the page comfortably legible on screen. Strong pushes both down and is intended for cases where the file simply has to fit an attachment limit and readability at a glance is enough. The honest guidance is to start at balanced, check the before-and-after figures the tool shows you, and step up or down from there. Because you can see both sizes before downloading, one pass at each setting takes seconds and tells you more than any general rule. If even the strong preset does not get you under a limit, splitting the document into two files is usually a better answer than degrading it further.
When to use a different approach entirely
If your PDF is mostly text and you need it smaller, rasterising is the wrong tool. Re-exporting from the original source document with image downsampling enabled, or using a PDF processor that recompresses embedded images while leaving text objects intact, will preserve selectable text and usually beat this method on size too. If you need the text to remain searchable — for accessibility, for archival, or because someone will need to copy from it — do not use the strong preset, and consider whether compression is necessary at all. Conversely, if your file is a photographed or scanned document, this approach is close to ideal and will typically outperform generic tools. One practical note: because the output pages are images, any fillable form fields, annotations, hyperlinks and embedded attachments are flattened away. Keep the original if any of those matter.
What compresses well and what does not
| Source document | Typical saving | Text stays selectable |
|---|---|---|
| Scanned pages or photos | 60–90% | No — was already an image |
| Presentation with large images | 50–80% | No |
| Mixed text and figures | 20–50% | No |
| Text-only, word-processor export | Little or none | No |
| Already-compressed PDF | Minimal | No |
Because pages are rasterised, output text is never selectable. Keep the original if you need searchable text.
Frequently asked questions
Is my PDF uploaded anywhere?
No. Rendering and re-encoding run entirely in your browser using pdf.js and pdf-lib. Nothing is transmitted or stored.
How much smaller will my PDF get?
It depends entirely on the source. Scans and image-heavy documents commonly shrink 60–90%. Text-only PDFs may barely change, because their text is already stored very efficiently.
Will the text still be selectable?
No. Pages are rasterised to images, so the output is picture-based. If you need selectable or searchable text, keep the original and compress a copy only for sharing.
Which compression level should I pick?
Start with balanced, look at the before-and-after sizes the tool reports, then try light or strong. Running all three takes seconds and is more reliable than guessing.
Why did my file get bigger?
That happens on text-only PDFs. Compact font instructions are being replaced with a picture of the text, which can cost more than it saves. Such files are not good candidates for this method.
Do form fields and links survive?
No. Flattening pages to images removes fillable fields, annotations, hyperlinks and attachments. Keep the original if you rely on any of them.
Why does compression take a few seconds?
Your device is doing all the work — rendering, downscaling and re-encoding every page. Longer documents naturally take proportionally longer.
What if it still is not small enough?
Try the strong preset, and if that is not enough, split the document into two files with the PDF splitter rather than degrading quality further.
Is this the best way to compress a text document?
No. For text-heavy files, re-exporting from the original source with image downsampling enabled preserves selectable text and usually produces a smaller file.
Last updated
Rendering via pdf.js, document assembly via pdf-lib, JPEG encoding via the HTML Canvas API. Compression characteristics reflect the rasterisation approach described above.