Compress a PDF
Reduce PDF size by downsampling the images inside it, without pretending every PDF can shrink by the same percentage. Ghostscript rewrites the document with object streams, compressed content streams and blank document metadata removed, and the actual byte result is reported rather than a promised percentage. PDFs built from large photographs may not become smaller because their image data is already compressed; PDFs with repetitive structure or unnecessary metadata have more room to improve.
Use this without the search next time. Prathom Workbench puts Prathom's tools in your toolbar.
Add to Chrome — freeWhat it does
- Optimizes PDF object and content-stream structure
- Removes blank document metadata fields
- Reports actual before-and-after bytes
- Preserves pages and embedded content
- Processes the file entirely in the browser
How to use PDF Compressor
- 1
Choose the PDF
Drop a PDF into the tool. The browser reads it into memory and does not send the document to a conversion endpoint.
- 2
Wait for the result
The document is parsed, rewritten with object streams, and saved as a new PDF. Large or complex files may take a moment in the browser.
- 3
Check the size change
Compare the before and after figures. A smaller result is useful; an unchanged or larger result means the source was already efficiently encoded.
- 4
Download the new copy
Save the compressed PDF and keep the original until you have opened and checked the pages you care about.
How it works
A PDF is a container of objects: pages point to content streams, fonts, images, annotations, and other resources. The compressor loads that object graph and saves it again with object streams enabled. Object streams pack small indirect objects together so their syntax costs less space, while the PDF writer applies its normal stream compression to content. Empty title, author, subject, keyword, creator, and producer fields are cleared from the new document.
The important boundary is image compression. A photo inside a PDF is already a compressed binary stream, and changing its size or quality would require a rasterizing and resampling policy. This tool does not make that destructive choice silently. That makes the output more predictable and private, but it also makes the result dependent on the source PDF's existing structure.
When it helps
Try it before emailing a text report, uploading a vector-heavy handout to a form, or storing generated PDFs that contain repetitive structure. Check the result on the receiving service because upload limits differ and a few kilobytes can matter. For photo-heavy scans, reduce the source images first or use an image-aware desktop workflow with a deliberate quality setting.
Compression is not encryption, redaction, or PDF repair. It does not remove visible confidential text, flatten a signature, recover a damaged cross-reference table, or make a scanned page searchable. Those are separate operations, and keeping them separate is how the output remains understandable.
Why a scan barely shrinks
If your file came out of a scanner or a phone document app, expect very little from this page, and the reason is worth knowing because it tells you what would work.
A scanned page is a photograph. There is no text in it, no fonts, no vector paths — just one large JPEG per page wrapped in enough PDF structure to display it. That JPEG is already compressed, and the object-stream packing this tool does applies to the PDF's own syntax, which in a scan is a rounding error next to the image data. Saving two kilobytes off a nine-megabyte file is the honest result.
Shrinking a scan means re-encoding its images, which means choosing a resolution and a quality level and accepting the loss. That is a decision with a visible consequence — text that was crisp becomes soft — so it belongs somewhere you can see the result before committing, not behind a button labeled "compress".
The practical routes: scan at 200–300 DPI in the first place rather than the maximum the hardware offers, choose grayscale for anything that is not a color photograph, and if the file already exists, use a desktop tool where you can set the downsampling target and compare pages. Where the goal is a smaller file and searchable text, OCR the scan to real text instead — the output is usually a fraction of the size and better in every other way too.
What actually gets smaller
The files that respond well to this page share a shape: lots of PDF structure relative to their content.
Reports generated by a library — invoices, statements, exports from a reporting tool — are the best case. They often contain thousands of small objects describing table cells, and object streams pack exactly that kind of repetition. Documents exported from word processors and slide decks with vector charts do well for the same reason.
Files that have been through several editing tools accumulate orphaned objects and redundant resources, and rewriting the object graph drops what is no longer referenced. If a PDF has been merged, split, and annotated over a year, there is often more to remove than anyone expects.
What will not move: a file that is already one image per page, a file produced by a tool that already writes object streams, and a file that is mostly one large embedded font. Run it and read the number rather than guessing — the operation is lossless, so the worst case is that nothing changes and you have lost nothing.
Examples
A text-heavy report
PDFs made mostly of text and repeated vector resources often benefit from object streams because the structural overhead is significant compared with the visible content.
A scanned photo packet
The source images already contain most of the bytes. A structural rewrite cannot remove photographic detail without resampling, which this privacy-first browser tool deliberately does not do.
Frequently asked questions
Does this compressor reduce image quality?
No image resampling is performed. The compressor rewrites PDF structure and metadata but does not turn a photograph into a lower-resolution image. That keeps the visible page content safer, but it also means a scan or photo-heavy PDF may not shrink much. Use an image-aware workflow when quality trade-offs are acceptable.
Why did my compressed PDF get larger?
A PDF that is already optimized may have little structural overhead left to remove. Rewriting it can add a small amount of object-stream bookkeeping or change how an existing producer packed resources. The tool reports that honestly instead of returning a larger file under a misleading “compressed” label.
Will the pages look exactly the same?
The operation preserves PDF pages and does not intentionally alter their visible text, vectors, or images. Still open the output before replacing an original, especially when the file contains forms, annotations, signatures, embedded attachments, or unusual viewer features that a general PDF library may not preserve perfectly.
Is my PDF uploaded?
No — and how it works changed on 2026-08-26. The file is sent to Prathom's own server, processed there, and deleted the moment the response is sent; nothing is stored and no account is needed. It used to run inside the browser tab, which was slower, capped by the memory that tab had, and — for several of these tools — produced worse output. Ghostscript downsamples the images inside the document, which a browser library could not do at all — the old version only re-saved the file and stripped metadata.
Further reading
- Flattening a PDF Made It 14 Times Bigger and Deleted Every Searchable WordFlattening is what you do when a form must stop being fillable or a signature must stop being movable, and it is often described as if it just locks things down. It does more than that: it replaces the page with an image of the page, and everything that made the page text goes with it.
- Converting a PDF to Grayscale Made It BiggerRemoving color sounds like removing data, so grayscale gets used as a compression step. On a document that was already black text on white, the conversion made the file larger — and the reason it did is the reason it sometimes makes a scan dramatically smaller.
- This PDF Kept Every Pixel and Still Came Out Looking WorstCompress a PDF and the obvious way to check what happened is to look at the resolution of the images inside. It is the number every tool reports and the one every comparison uses. It is also, on its own, close to useless — here is a file that kept every single pixel at full resolution and looked worse than one that threw three quarters of them away.