PDF Compressor
Compare PDF object cleanup with JPEG page compression, entirely locally
About this tool
PDF Compressor offers object cleanup and JPEG page rebuilding. Cleanup removes all document Info entries, including custom keys, and the catalog XMP reference before rewriting object streams. It preserves page content and does not recompress embedded photographs or erase every possible metadata location. JPEG mode renders displayed pages at a chosen scale and quality, optionally in grayscale, while keeping their physical CropBox dimensions and rotation. Existing supported form appearances become pixels; other annotations, links, text selection, attachments and accessibility structure are discarded. Neither method guarantees a smaller file: the result reports the exact input and output sizes, including growth. Processing is local and downloads are explicit. Files are limited to 50 MiB and 200 pages; image output is limited to 40 pages and bounded pixel counts. XFA, encryption and any signature fields are rejected.
Scenario Recipes
Choose a smaller, usable copy
Goal: Compare actual file size without sacrificing a required PDF feature.
- Try the example, then choose your own unsigned PDF and start with object cleanup.
- If page images are acceptable, choose JPEG mode and a render scale and quality; use grayscale only when color is irrelevant.
- Download the result and compare page dimensions, text needs and file size. Keep the original if the result grows or loses a needed feature.
Result: A measured compression decision with a separate output file.
Failure Clinic (Common Pitfalls)
The size did not decrease
Cause: The source may already be optimized.
Fix: Keep the original or compare a lower JPEG quality if image output is acceptable.
A note or hyperlink disappeared
Cause: JPEG mode intentionally discards non-form annotations and interaction.
Fix: Use cleanup mode when those features must remain; verify the downloaded copy.
Production Snippets
Interpret a size comparison
text
Input: 431 B
Output: 546 B
Change: 26.7% larger
Decision: keep the original when no other change is needed.Frequently Asked Questions
Which mode should I use for a digital PDF?
Start with object cleanup when selectable text, vectors and form structure matter. It rewrites the existing PDF without rebuilding pages as images. Compare the result before replacing the original.
Why did my PDF become larger?
Object streams and serialization have overhead, and embedded images are not recompressed in cleanup mode. JPEG rebuilding can also grow a small digital PDF. Growth is shown as a positive larger percentage; keep the original when the output is less useful.
Does render scale change the paper size?
No. Scale controls pixel detail. Each output page uses the source displayed CropBox width and height, including rotation; increasing scale does not enlarge the physical sheet.
What does JPEG mode discard?
The output contains page images. Supported existing form appearances are included in those pixels, but ordinary annotations, links, selectable text, attachments and accessibility structure are discarded. JPEG images remain editable in an image or PDF editor.
Does object cleanup erase private data?
No. It removes the document Info and catalog XMP references. Page XMP, attachments, image metadata, annotations and unused object bytes can remain. Use the dedicated metadata tool to choose additional XMP references, and a specialist sanitizer for forensic removal.
What limits and unsupported PDFs apply?
Input is at most 50 MiB and 200 pages. Image output supports 40 pages, 8,192 pixels per side, 16 million pixels per page and 80 million in total; output is limited to 80 MiB. Encrypted PDFs, XFA and signature fields, including empty ones, are rejected.
Keep browsing