How to compress a PDF (free, no upload)

Why PDFs get large

Most oversized PDFs fall into one of three categories:

• Scanned documents. A scanned page is an image. At standard scanner resolution (300 DPI), each page is a 2–5 MB image before any compression. A 20-page scanned contract can be 40–100 MB before you've done anything with it.
• Exported from design tools. PDFs exported from Illustrator, InDesign, or Figma often include high-resolution embedded images at full print quality (300+ DPI). A web-ready PDF rarely needs more than 72–150 DPI for the embedded images.
• Embedded fonts. PDFs can embed entire font files to ensure text renders correctly anywhere. Some fonts are 1–2 MB each. A document using several custom fonts can be large before any images are considered.

Most PDF compression tools work by re-encoding the embedded images at lower resolution or higher compression. The text and structure of the document stay intact.

Method 1: browser-based compression (fastest)

Drop the PDF into a browser-based compressor. It re-encodes the embedded images using JavaScript, produces a smaller file, and downloads it. Your file never leaves your device.

This works well for PDFs where images are the main size driver. The limitation is that browser-based tools have less control over the exact output than server-side tools running tools like Ghostscript.

Good for: quick compression of any standard PDF, documents with scanned pages, anything with embedded photos.

Method 2: online upload tools

Tools that process files on their servers (using Ghostscript or similar) often produce smaller outputs than browser-based tools, especially for complex PDFs. The trade-off is that your document leaves your device.

For most documents, the privacy concern is minimal. For contracts, financial statements, medical records, or anything with personal information, think twice about uploading to a free service you don't have a relationship with.

File size limits are also common on free tiers of upload-based tools: typically 10–25 MB per file, which is a problem for large scanned documents.

Method 3: reprint from the source

If you have access to the original file (Word doc, Figma design, Keynote presentation), the most effective compression is to re-export the PDF with optimized settings.

In most applications, the PDF export dialog has a quality or optimization option. "Screen" or "Web" quality uses 72–96 DPI for images. "Print" quality uses 300 DPI. For anything that will only be read on screen or emailed, screen quality produces much smaller files with no noticeable difference.

This approach won't help for PDFs you received from someone else and don't have the source for. In that case, use method 1 or 2.

Compression results by document type

Step-by-step with Filagram

1. Open the PDF compressor below.
2. Drop your PDF onto the drop zone, or click to select it.
3. Choose a compression level. "Medium" (the default) works for most use cases. "High" produces smaller files with slight quality reduction in embedded images.
4. Click Compress and download the result.

Processing happens in your browser. A 10 MB PDF typically takes 2–5 seconds on a modern device. The original file is not modified.

When compression won't help much

If your PDF is mostly text with very few images (a contract, a report, a form), compression tools won't produce dramatic results. The file size is driven by fonts and text data, which don't compress as aggressively as images. For these documents, 20–40% reduction is more realistic.

Password-protected PDFs generally can't be compressed without removing the password first.