How browser file processing works
What “processed on your device” means on ConvertImgVidPDF, and what it does not mean.
The selected file can stay local
When you choose a file, browser APIs can read it into memory and JavaScript or WebAssembly can transform it without posting the file contents to a remote conversion endpoint. That is the normal design for the current tools.
The website still uses the network
Loading a page, stylesheet, JavaScript module, icon or WebAssembly engine creates ordinary web requests. Local file processing refers specifically to the selected file contents, not to the absence of all network activity.
Heavy engines are loaded only where needed
PDF rendering uses PDF.js and media conversion uses FFmpeg WebAssembly. Those heavier assets are requested only on pages that need them; ordinary image and data tools do not automatically load the media engine.
Browser memory is the main practical limit
Large PDFs, archives and videos may require far more memory than their compressed file size. Phones can run out of memory earlier than desktop browsers, and long media conversions can consume significant CPU and battery.
Outputs should be verified
Conversion can change quality, metadata, page structure, codecs or compatibility. Keep important originals and open the generated result before using it for a submission, archive, legal document or production workflow.