JoyToolsJoyTools

PDF Compress

100% Client-Side Private — Zero Server Uploads

Reduce the file size of your PDF documents by stripping metadata and optimizing internal structures.

🔒 100% Client-Side: All data is processed locally in your browser and never leaves your device (Zero Server Uploads).
E-E-A-T Security Proof →
PDF Compressor
100% Client-Side Private — Zero Server Uploads

Reduce the file size of your PDF documents with customizable compression levels. Choose 100% private browser compression or deep stream optimization on our Cloud Engine.

Drop a PDF file here to compress

Supports single PDF files up to 300MB (or click to browse)

1. Compression Level

2. Processing Engine

💡 Recommendation for Selected Level:

Best choice for 95% of documents. Significantly reduces file size for fast email and web sharing while maintaining sharp image & text quality.

Share & Developer Embed

How do I use PDF Compress?

1

Drop your PDF file into the compressor or click "Load Sample Data" to test with a sample document.

2

Choose a compression preset (`Extreme / 72 DPI`, `Recommended / 144 DPI`, or `Lossless Structure Optimization`).

3

Compare the before/after byte savings percentage and download your compressed PDF.

What features does PDF Compress offer?

Smart multi-level compression presets: Recommended (150 DPI), Low (Print quality), and Extreme (Email ready)

Selective image stream downsampling that retains 100% sharp vector typography and line drawings

Removal of duplicate font descriptors, unused color profiles, and redundant uncompressed metadata objects

100% client-side in-browser WebAssembly execution or 300MB high-capacity Cloud Engine

Instant real-time file size savings calculator and interactive percentage progress tracking

Zero file data retention policy guarantees confidential financial and legal files never leave your memory

PDF Stream Filters, DPI Downsampling & Byte Reduction Benchmarks

Compares lossless object-stream cleanup (`FlateDecode`) against lossy raster downsampling (`DCTDecode` JPEG / `JPXDecode`).

PDF Compression Presets: Target Resolution, Stream Filter, and Size Reduction
Compression PresetTarget Image DPIStream Encoding FilterTypical Size ReductionBest Use Case
Lossless / StructuralNative (Unchanged)FlateDecode (zlib) + Orphan Object Purge5% – 20%Vector CAD drawings, contracts,LaTeX papers
Balanced (Recommended)144 – 150 DPIDCTDecode (JPEG Q=75) + Flate45% – 70%Email attachments, reports, slide decks
Extreme (Web / Portal)72 – 96 DPIDCTDecode (JPEG Q=55) + Metadata Strip70% – 90%Government portal uploads with strict 2MB limits

Frequently Asked Questions about PDF Compress

Why did my text-only PDF barely shrink under Lossless compression?

Text and vector paths inside modern PDFs are already compressed with `FlateDecode` (`zlib` DEFLATE). Significant size reductions (`50%–85%`) occur primarily when embedded high-resolution raster images (`300+ DPI`) are downsampled to `144 DPI` or `72 DPI`.

What is the difference between `FlateDecode` and `DCTDecode` inside a PDF?

`FlateDecode` is lossless LZ77+Huffman compression used for text streams, fonts, and vector paths. `DCTDecode` is Discrete Cosine Transform (JPEG) lossy compression applied specifically to continuous-tone `/Image` XObjects.

Does rasterizing/downsampling a PDF affect selectable text?

Structural optimization preserves 100% of selectable text and vector fonts, whereas full-page canvas rasterization flattens pages into compressed JPEG/WebP streams to guarantee maximum file size reduction for scanned documents.

How does stripping `/Metadata` and incremental-update (`xref`) history save space?

PDF editors often append changes as incremental updates at the end of the file, leaving deleted images and XMP metadata blocks orphaned in the byte stream. Rebuilding a clean single-pass `xref` table purges all dead objects.