Why your PDF is still large

Take it step by step

Check whether text can be selected. A page-sized picture usually needs a different approach from a text document.
Inspect image-heavy pages and remove unnecessary attachments or pages.
Try a lossless pass first when text, forms or links must remain.
Compare the output size with the original before choosing any lossy pass.

A CONCRETE EXAMPLE

A situation to try.

A one-page scanned letter can be larger than a ten-page text report. Downsampling that scan may help; rasterizing the text report can make it worse.

Use the tool’s synthetic example first where available. The example here explains a decision; it is not a reported benchmark result.

Before you call it done.

Check attachments before discarding them.
Compare output bytes and page count.
Verify form fields and text search.

Know the limits.

Compression cannot reliably make every already-optimized PDF smaller.

This tool’s current boundary: Lossless reserialization only in this build. Image recompression is not available. Already optimized files may not shrink.

MEASURED, NOT ASSUMED · 2026-10-05

Our lossless compression check.

Three synthetic PDFs were processed through the actual compression adapter. PDF.js independently read and rendered all five input/output pages. Extracted text and rendered pixels matched at scale 1 in this Windows/Node test.

Internal fixture results — not a competitor ranking
Fixture	Input bytes	Output bytes	Saved	Text layer
cv	1,823	1,167	35.98%	Retained
scan	8,670	8,241	4.95%	Absent before and after
brochure	27,538	25,871	6.05%	Retained

The source fixtures used uncompressed object streams. These reductions do not predict results for already-optimized files. Forms, signatures, attachments, accessibility and browser/mobile performance were not certified by this check.

Download and reproduce the evidence

Versions, method and SHA-256 hashes (JSON) · Raw results (CSV)

Reproduction: run npm test -- tests/tool-evidence.test.ts in the project checkout. The downloadable report records the environment. Timing is a single Node adapter run, not a browser speed score.

Evidence and scope

Technical source guidance and task-specific output checks. These do not establish that every file, browser or device passes.

Use the checks above on your own exported result. A source explains the format or mechanism; it is not proof that this export preserved every feature.

Technical reference: PDF-LIB: page operations and exports ↗

How we verify outputs · Device compatibility