Compress PDF
Drop a PDF, get a leaner version with pages, fonts and images untouched.
Drop your file here
or click to browse — up to 1024 Mb.
How to compress a PDF
Why compress a PDF?
Oversized PDFs run into limits everywhere. A signed contract bounces off a 10 Mb email cap. A portfolio is too heavy for the upload form. A court e-filing system rejects an exhibit over its size limit. A year of invoices quietly bloats your cloud storage. Compressing the file makes it fit, upload faster, and cost less to keep — without you having to rebuild anything.
This tool rewrites the file losslessly. It deflates every content stream with zlib, drops unused objects and duplicated resources, then emits a fresh, compact cross-reference table. Fonts, text and images come out bit-for-bit identical — nothing is re-encoded or degraded, so the compressed PDF looks exactly like the original. It runs in a Rust service over an encrypted TLS connection, your file is auto-deleted the moment the result is returned and is never shared, and a content-hash cache makes repeat uploads instant. No watermark, no signup, no upsell.
What actually makes a PDF big
Not all PDFs are heavy for the same reason, and that determines how much compression can help.
• Scanned pages — by far the most common cause. A scan is a full-page photograph, and at high resolution each page can be several megabytes. This is image data, not text.
• Uncompressed streams — Word, LibreOffice, and many export pipelines write their page content without compressing it. This is pure waste, and it's exactly what lossless compression removes.
• Embedded fonts and duplicated resources — the same logo or font subset stored on every page instead of once. Deduplication reclaims that space.
• High-resolution photos — a print-quality image at 300 DPI carries far more data than a screen needs, but shrinking it is a lossy choice this lossless tool leaves untouched.
Lossless compression vs downsampling
There are two fundamentally different ways to shrink a PDF, and it's worth knowing which one you're getting. Lossless compression — what this tool does — removes redundancy: it packs uncompressed streams, deletes unused objects, and rebuilds the file structure. Every pixel and glyph survives exactly, so there is zero visible change. The trade-off is that it can only remove waste, so a file that was already efficient barely moves.
Downsampling is the other approach: it lowers the resolution of embedded images, throwing away pixel detail to make the file dramatically smaller. It's the only thing that meaningfully shrinks a heavy scan, but it is lossy — zoom in and the image is softer. For most documents you want the lossless pass first; reach for downsampling only when a scan is still too big and you can accept slightly less sharp images.
When compression helps — and when it won't
| Pdf Type | Expected Gain | Why |
|---|---|---|
| Word / LibreOffice export | High (30–60%) | Content streams ship uncompressed |
| Slide deck exported to PDF | Medium–high | Repeated images and fonts to dedupe |
| Plain text PDF | Low | Text is tiny and often already packed |
| Already-optimized PDF | Very low | Little redundancy left to remove |
| High-resolution scan | Low (lossless) | Bulk is image data; needs downsampling |
Compressing to beat an email or upload limit
Most mailboxes cap attachments around 10 to 25 Mb, and upload forms are often stricter. If a compressed file still doesn't fit, you have two good options that don't touch quality. Split the PDF into parts and send them separately, then the recipient merges them back — useful for long documents. Or, if the weight comes from scanned images, accept a lossy downsample: it's usually the difference between a 40 Mb scan and a 4 Mb one. For born-digital documents (contracts, reports, invoices), the lossless pass here is almost always enough on its own.
Frequently asked
Is it free?
Yes, no signup, no watermark, no daily quota. Every compression is free.
How much smaller will my file be?
Usually 15–60% smaller. Office exports (Word, LibreOffice) shrink the most because they ship uncompressed streams; already-optimized files and scans gain less.
Will the quality drop?
No. This is lossless compression — we only rewrite streams and the cross-reference table, so fonts, text and images are preserved bit-for-bit.
Why didn't my file get much smaller?
Because it was already efficient. A text PDF or a scan that was already downsampled has little redundancy left to squeeze. Lossless compression can only remove waste, not re-encode images.
My scan is still huge — what now?
Scans are large because they are full-page images. Lossless compression helps a little; the real lever is reducing image resolution, which trades some sharpness for a much smaller file. That's a separate, lossy step.
What's the max input size?
50 Mb per file. Split larger documents first, compress the parts, then merge them back if needed.
Are my files stored?
No — files travel over an encrypted TLS connection, are processed in memory, and are auto-deleted the moment the result is returned.
Does it work on scanned PDFs?
Yes, but expect smaller gains: the bulk of a scan is already-encoded image data, and lossless compression leaves those pixels untouched.