How the PDF Compressor Works
DocDox PDF Compressor reduces PDF file sizes by recompressing embedded images using JPEG DCT (Discrete Cosine Transform) encoding, all within your browser via WebAssembly. Most PDF bloat comes from embedded photographs and scanned pages — this tool targets exactly those elements while leaving text, vector graphics, and document structure untouched.
When you upload a PDF, a background Web Worker analyzes each embedded image using OffscreenCanvas, converts it to JPEG at your chosen compression level, then injects the smaller version back into the PDF structure using pdf-lib. Three profiles are available: Low (preserves visual quality, modest size reduction), Recommended (best balance for most documents), and Extreme (maximum compression for size-critical situations).
Because processing happens locally, large PDFs that would time out on server-based services work here. There are no upload limits, no queue times, and no files stored on any server after your session ends.
Does compressing a PDF affect its text quality?
No. Only embedded images are recompressed. Text, fonts, and vector graphics are completely unaffected.
How much can file size be reduced?
PDFs with many high-resolution photos can be reduced by 50–80%. PDFs that are mostly text may see only modest reductions.
Is my PDF sent to any server?
No. The entire compression process runs in your browser using a Web Worker and WebAssembly.