When this tool is useful
Start with the conservative mode for PDFs whose text, links and forms must remain usable. Use scan mode only when a flattened visual copy is acceptable.
How local processing works
The default mode rewrites supported PDF objects without rendering pages to images, then reopens the result for structural checks. Scan mode instead renders each page as JPEG at the selected resolution.
How to use it
- Choose the source files.
- Review the available settings and page order.
- Process locally, inspect the result, and download it.
Important limitations
No size reduction is guaranteed. Conservative mode rejects encrypted, signed, XFA and unsupported documents. Scan mode removes searchable text, vectors, links, interactive forms and accessibility. If the result is not smaller, keep the original.
Verify the downloaded result
Reopen the downloaded file in its destination application, check dimensions or page count and visible content, and keep the source until the result has been accepted.
Private by design
Your files stay in this browser. The tool does not upload the source or output to our servers.
Questions and answers
Does this tool upload my files?
No. Processing happens in your browser; closing the page clears the working session.
Will the output always match the source exactly?
No size reduction is guaranteed. Conservative mode rejects encrypted, signed, XFA and unsupported documents. Scan mode removes searchable text, vectors, links, interactive forms and accessibility. If the result is not smaller, keep the original.
What should I check before using the result?
Reopen the downloaded file in its destination application, check dimensions or page count and visible content, and keep the source until the result has been accepted.
Which browsers and devices are supported?
Use a current browser with JavaScript, workers, canvas, WebAssembly where the format requires it, Blob downloads, and enough memory. Reopen important downloads in the destination application.
Concrete input and output example
Try a 12MB scanned PDF in conservative mode first and compare the output bytes. If a flattened copy is acceptable, explicitly choose scan mode and compare zoomed text. Neither mode promises savings; a result that is not smaller returns the original bytes.
Browser compatibility notes
Reviewed against the implementation and required web APIs on 2026-09-02. Use a current browser where these APIs are available and verify important output on your device.
- Chrome / Edge (current desktop)
- Requires local PDF parsing, worker/canvas APIs where applicable, document creation, and Blob download.
- Firefox (current desktop)
- Uses the local PDF workflow; always verify the generated download in its destination viewer.
- Safari (current desktop)
- Uses the same local workflow; large page sets can have a lower practical memory ceiling.
A tool-specific question
Does compression turn text and forms into images?
Not in the default conservative mode: supported objects are rewritten and checked without rasterizing pages. The optional scan mode does remove searchable text, links and interactive forms. Complex unsupported PDFs are rejected, and neither mode guarantees a smaller file.