PDF Bulk PDF compress
Queue a folder of PDFs, compress every one of them with the same settings, and take the whole batch away as a ZIP.
Drag files here, or
Add as many as you like. Every file is compressed in this tab, so confidential documents never leave your device.
Ready. Runs locally on your device.
Your files stay on your device. The tool works directly in your browser, using your device to process your files. Nothing is sent to our servers, and we never receive, store, or see your files or figures.
Share result
Cite
Cite this page
Reembun. (2026, August 24). Bulk PDF compress. https://reembun.com/bulk-pdf-compress
How to use it
Queue every file
Drop a whole folder of PDFs on the box, or press Choose PDFs and select them all. Each one is opened as it arrives, so a damaged or locked file is reported before the run starts rather than during it.
Pick one method for the batch
Keep text sharp rebuilds each file without touching its text. Shrink hard redraws the pages as images, which is far smaller on scans but turns text into pixels. The setting applies to everything in the queue.
Press Compress
Files run one after another with a progress bar each, and a running total of what has been saved. Press Stop at any point and everything already finished stays where it is.
Take the results
Save any single file from its own row, or use the download button to take the whole batch as one ZIP.
What a batch changes
Compressing one PDF is a thirty second job that needs no special tool. Compressing forty is a different task, and it usually arrives in one of these shapes:
- A folder of scans that has to fit an upload limit, page by page.
- A month of invoices or receipts headed for an accountant.
- Every attachment for one application, each with its own size cap.
- An archive of documents that has quietly grown to several gigabytes.
All four share the same shape. The decision is made once, the work is repetitive, and the expensive failure is losing an hour of it near the end.
One decision, applied to everything
The method, the resolution and the quality apply to the whole queue. That is deliberate. Setting them per file would turn forty documents into forty decisions, which is the thing you came here to avoid.
Keep text sharp (lossless)
Each document is parsed and rewritten with compressed object streams, and its metadata (title, author, producer, keywords) is stripped. Page content is untouched, so text stays selectable and images keep their quality. Typical saving is 5% to 20%, and sometimes nothing at all on a file that was already well produced.
Shrink hard (image based)
Every page is rendered to a bitmap at the resolution you choose, compressed as JPEG, and placed into a new PDF at the original page dimensions. On scans this routinely saves 70% to 90%. The cost is real: text becomes part of a picture, so it can no longer be selected, searched, copied or read aloud by a screen reader.
For a mixed queue, sort before you run. Scans belong in image mode, and anything whose text matters as text belongs in lossless mode.
Reading the numbers
Each row shows its own before and after size with a percentage, and the total above the list shows what the batch achieved as a whole. Both matter for different reasons. The total tells you whether the run was worth doing, and the per file figures tell you which documents were carrying the weight.
| What dominates the file | Lossless saving | Image mode saving |
|---|---|---|
| Scanned pages | Almost none | 70% to 90% |
| High resolution photographs | Almost none | 60% to 85% |
| Text with embedded fonts | 10% to 20% | 40% to 70%, and the text is gone |
| Vector charts and diagrams | 5% to 15% | Often worse, because vectors compress better than a picture of them |
That last row is the one that surprises people. A batch of diagram heavy PDFs run through image mode comes out larger and blurrier at the same time.
Why the files run one at a time
Rasterising a page holds a full size canvas in memory while it is encoded. Running several documents in parallel would hold several of those at once, and that is precisely how a browser tab dies two thirds of the way through a job it had nearly finished.
Sequential is slower and it finishes. On a batch tool that is the only quality worth optimising for. Stopping is honest for the same reason: press Stop and the file in progress is abandoned rather than written out half built, while everything already compressed stays in the list ready to download.
Order of operations for an upload
When a form wants one file under a hard limit, the sequence that works is:
- Convert anything that is not yet a PDF, with Word to PDF for documents.
- Merge the parts into one document in the order the form asks for.
- Compress the merged file once, in image mode if it is mostly scans.
Compressing each part first and merging afterwards is the common instinct and it produces a larger file, because every source document brings its own copy of the fonts it embeds.
When the form wants separate files under separate limits, that is when this page is the right tool: same settings, one press, one ZIP.
Metadata comes off too
Lossless mode clears the title, author, subject, keywords, producer and creator fields on every file in the queue. Those fields leak more than people expect, routinely including the original author’s name, the software that produced the file and sometimes an internal path from the machine it was made on.
Note that this is document metadata. It does not touch the contents of the pages. Text hidden behind a black rectangle is still text, and it is still in there.
Privacy
Nothing is uploaded. Every document in the queue is opened, rewritten and zipped inside this tab, which is the whole reason there is no file count limit here. Contracts, medical records and bank statements can go through a batch of forty without a single byte leaving the machine.
Common questions
How many PDFs can I queue at once?
There is no number written into the tool, because there is no server to protect. The limit is your own device. A laptop handles a few hundred megabytes of documents without complaint, and a phone with little free memory will start to struggle sooner. Files are compressed one after another rather than all at once, which is exactly what keeps a long queue survivable on modest hardware.
Can I use different settings for different files?
Not in one run. One method, one resolution and one quality apply to everything in the queue, because that is what makes a batch a batch. If your files genuinely need different treatment, group them. Run the scans together in image mode, clear the queue, then run the text documents in lossless mode.
What happens if one file fails?
It is marked with the reason on its own row and the queue carries on. When the run ends you can remove that file and press Compress again, which only picks up the ones that have no result yet. A single bad file never costs you the batch.
Why did some files get bigger?
Because they were already stored more efficiently than the setting you chose. Rasterising a vector chart at 200 dpi produces more data than the vectors it replaced, and re-encoding an already aggressive JPEG scan can do the same. The per-file percentages make this visible immediately, so keep the originals for those and rerun the rest at a lower resolution.
Does compressing each file separately beat merging first?
For a submission that wants one document, merge first and compress once. Compressing the parts and then merging them leaves you with every embedded font subset duplicated across the sections, and the combined file ends up larger than a single compression pass would have made it.
Are the files uploaded anywhere?
No. Every document is opened, rewritten and packed into the ZIP inside this browser tab. Nothing is transmitted, which is why there is no queue to wait in and no retention window to trust.
Last reviewed
