PDF · Compress

Compress PDF

Strip unused objects and recompress object streams. Best on PDFs with redundant data — limited on heavily photo-based PDFs.

⌗
Drop a PDF here or click to choose
Single PDF
Honest about limits: browser-only compression is structural — it removes orphaned objects and uses object streams. Expect 5–15% on photo-heavy PDFs and up to 50% on text-only PDFs with lots of metadata. For aggressive photo recompression you need a server-side tool.

Compress PDF in your browser — honest, fast, private

Whenever a website says "compress your PDF for free," you're usually being asked to upload it. That's how the big PDF sites work — they receive your file, run server-side optimizations like JPEG re-encoding, and send back a smaller version. The trade-off: your potentially sensitive document sits on their servers, often for hours.

Toolvana takes the other path. Our compression runs entirely in your browser using pdf-lib. It can't match server-side photo recompression on the biggest file-size wins (we'd need a WASM JPEG encoder for that — coming soon), but it does the structural work cleanly and never touches the network.

What "compress" actually does here

PDFs accumulate a lot of unused data over their lifetime: orphaned objects, redundant resource dictionaries, oversize metadata, uncompressed object streams. Our tool walks through every object, removes unreferenced ones, packs the rest into compressed object streams, and rewrites the cross-reference table. For text-heavy PDFs (contracts, reports, ebooks), this often delivers 30–50% reductions. For photo-heavy PDFs (scanned documents, image-rich brochures), expect 5–15%.

When to use a different tool

If your PDF is mostly photos and you need to drop it from 50 MB to under 5 MB for email, you'll get better results from a server-side tool that re-encodes the embedded images at lower JPEG quality. That's the brute-force win our browser tool can't match yet. Honest about it — better that than a misleading "92% compression" promise.

Use cases where this tool excels

  • Cleaning up PDFs exported from old desktop software (Word, LibreOffice, InDesign) that bloat with metadata
  • Stripping unused fonts and color profiles from PDFs you're about to email
  • Removing image-resource duplicates that accumulate when you merge PDFs from different sources
  • Privacy-sensitive documents (medical, legal, financial) where uploading to a third party is a non-starter

FAQ

Why didn't my photo PDF compress much?

The JPEG images inside are already compressed. To shrink them further, you'd need to re-encode at lower quality — that's a feature on our roadmap (mozjpeg.wasm), not in the current version.

Is the output safe to use?

Yes. The compressed PDF is a valid PDF that opens in every viewer. We only remove unused data; the visual content is bit-identical to the original.

Does the file leave my browser?

No. Open DevTools → Network tab while compressing; you'll see zero upload requests. Everything happens locally.

How big a PDF can I compress?

RAM-bound. Modern desktop browsers handle 500 MB+ without trouble. Mobile devices: roughly up to 100 MB.

Does compression hurt searchability?

No — the text layer is preserved. Search inside the compressed PDF still works.