PDF Compressor

• Fully client-side. The PDF is read with the File API, parsed by pdf-lib, and its images are re-encoded with the HTML5 Canvas API. No content is transmitted to any server.
• No tracking. The tool does not record which files you upload or what settings you choose. There is no analytics on file content.
• No automatic saving. Files are held in memory only. Reloading the page clears everything.
• Compression is not always possible. PDFs that are already optimised, or that contain mostly text and vector graphics, may not shrink — and in rare cases may grow slightly. The tool tells you when this happens so you can keep the original.
• Text and vector content is preserved. Text stays selectable and vector shapes stay sharp. The re-encoding step applies only to embedded raster images.
• Encrypted PDFs are not supported. If a file requires a password, remove the password first and save an unprotected copy.
• Memory limits. Files up to about 50 MB work well. Larger files may be slow because all processing happens in memory.

A practical guide to PDF compression

PDF is a container format. It stores text, vector graphics and raster images side by side, along with the metadata that ties them together. That flexibility is why PDFs are so widely used, and it is also why some PDFs are huge and others are tiny. A twenty-page text document may weigh less than 200 KB; a ten-page brochure with high-resolution photographs may weigh 30 MB. Compressing a PDF is mostly about finding where the bulk lives and reducing it without visible loss.

Where the bytes come from

Three sources account for almost all of a PDF's file size:

• Embedded images. Photographs, screenshots and scans are almost always the dominant cost. A single 300 dpi A4 scan can be 5–15 MB on its own.
• Embedded fonts. Fonts are usually a few hundred kilobytes each, and a document that uses six typefaces embeds all six. Subsetting helps, but the fonts are still substantial.
• Structure and metadata. The page tree, cross-reference table and XMP metadata add up to a few kilobytes even in a simple document, and can grow with heavy annotation.

Text itself takes almost no space. It is the images that make PDFs heavy, which is why the most effective compression is image re-encoding.

Lossy vs. lossless image compression

Raster images inside a PDF are typically stored as either JPEG (lossy) or Flate/PNG (lossless). Re-encoding a lossless image to JPEG at moderate quality usually produces a file 5–10× smaller with no visible difference on screen. Re-encoding a JPEG that is already compressed at high quality produces a smaller file but with more visible artifacts, which is why good tools offer a range of quality settings rather than a single "compress" button.

What this tool does

This compressor renders each page in a temporary canvas, extracts the embedded raster images, downsizes them if they exceed the maximum width you chose, re-encodes them as JPEG at your quality setting, and rebuilds the page using the newly encoded images. Text and vector graphics are left untouched, so they stay crisp and selectable. The result is a new PDF, and the original file on your disk is not modified.

When compression is limited

Not every PDF can be made smaller. If the document is mostly text, if its images are already small, or if it has already been through a professional optimiser, there may be little room to save. In a few edge cases the re-encoding step can even produce a slightly larger file than the original. The tool is transparent about this: it shows you the exact sizes before and after, and it tells you when the result is larger so you can keep the original.

About this tool

This PDF compressor is part of the TempScript free online tools collection. It runs entirely in your browser, requires no account, and never sends your files anywhere. If you work with PDFs regularly, you may also find our PDF merger and PDF page organizer useful.