How Local Browser-Based File Processing Works

Clicking Choose Files or dropping a file onto a tool hands that file to the page as data your browser already has in memory — nothing is sent anywhere at that moment. This is the actual difference from a typical "upload and convert" website: there, choosing a file is normally followed by the browser sending its bytes to a remote server before any work starts. Here, that sending step doesn't happen for these tools — the file is read and processed using your browser's own features, on your own device.

"Runs in your browser" covers more than one mechanism, depending on the tool:

Canvas-based tools — like Image Compressor and PDF Image Extractor — decode your image into an invisible drawing surface called a canvas, built into every browser, then read the result back out as a new file. This is the same feature web pages use to draw graphics; here it's used to resize, recompress or re-encode a picture without a server.

JavaScript library tools — like PDF Text Extractor — use a JavaScript library (PDF.js) running in a background worker to read a PDF's internal structure and pull out its stored text, page by page, without ever drawing the page or leaving the main page unresponsive while it works.

WebAssembly-based tools — like MP4 to GIF Converter — run a real program compiled to WebAssembly, a format browsers can execute at close to the speed of a normal desktop application. MP4 to GIF uses one that is a real video encoder, downloaded the first time you use it — about 30 MB — and cached by your browser afterward, which is exactly why converting video feels heavier than the other tools: it genuinely is running full video encoding, not a shortcut.

Once a tool finishes, the result exists in memory as a Blob — a chunk of data with a type attached, such as "this is a PNG image." To show you a preview or let the Download button save it, the page creates a temporary Object URL that points at that Blob. That URL only works inside your own browser tab; it isn't a real address anywhere else, and it's released as soon as you replace the file, remove the result, or leave the page. Nothing is written anywhere permanent unless you press Download yourself.

Because the file itself never has to travel to a conversion server for these tools to work, it is not exposed to that server as part of processing it. That is a specific, narrow claim about the file you select — it is not a claim that the website makes no network requests at all. Loading the page's own code, and for some tools the JavaScript library or WebAssembly program itself, are ordinary requests like any website makes; they carry the program, never your file's content.

Local processing has real limits, too. Everything runs using your own device's memory and processing power, so a very large file or an older, lower-memory device can make a tool slow, or in extreme cases cause the browser tab to run out of memory — there is no server to quietly take over if that happens. And because nothing is kept on a server, closing or reloading the tab means anything you haven't already downloaded is simply gone; there's no account or history to recover it from afterward.