"Your files never leave your device" is an easy sentence to put on a web page. This article is about checking it, because most free online tools do the opposite of what that sentence says: they send your file to a server, process it there, and send the result back. For a meme it does not matter. For a signed contract, a passport scan or a client's unreleased artwork, it does.

Below: what "no upload" actually means, a test anyone can run in five minutes, what I found when I ran it on the tools on this site, and the list of those tools grouped by job.

What "no upload" means

A client-side or in-browser tool does the processing on your own computer, in code that runs inside the browser tab. The browser reads the file you picked, resizes or re-encodes or counts it, and hands you a download. The network is never involved after the page loads.

A server-side tool sends the file over the network to the provider, processes it there, and returns the result. The two can look identical on screen. The path the file takes is completely different, and so are the consequences:

  • Privacy. No server copy means nothing to log, retain, analyse or leak.
  • Speed. No upload-and-download round trip, which matters on slow or metered connections.
  • Offline. Once the page has loaded, a true client-side tool keeps working with the Wi-Fi off.
  • No account. There is nothing to store, so there is nothing to sign up for.

The five-minute test

Every desktop browser has a Network panel that lists each request a page makes, with its size. Mozilla documents it for Firefox under Network Monitor; Chrome and Edge have the same panel.

  1. Open the tool and let it finish loading.
  2. Press F12 and open the Network tab.
  3. Clear the list, then add a test file with nothing private in it.
  4. Run the tool and watch for new requests.
  5. An upload shows up as a large POST or PUT request about the size of your file. If nothing of the sort appears, the work stayed on your device.

Then the second half: with the page still open, switch off your internet connection and run the same task again. If it still works, there is no server in the loop. Reloading the page may fail while you are offline, because the page's own files are no longer reachable; test the page that is already open.

Two cautions. A page can load its libraries on demand, so run the task once while online before you test offline. And a site can be local for one job and not another: image compression may run in the browser while OCR on the same site calls an API. Test the task you care about.

What I ran it on

I added a public JPG of 63.9 KB to the DocFlow PDF tools on this site. It appeared in the Images to PDF list with its name and dimensions, 1200 × 630, and the Create PDF button, the A4 page option, orientation and margin all became active with no account. The conversion runs in the browser: the page's code makes no request that carries your file, and the Network tab test above lets you confirm that for yourself.

The Images to PDF tool on this site, before a file is added: the drop zone, then page size, orientation and margin.
The Images to PDF tool on this site, before a file is added: the drop zone, then page size, orientation and margin.

The tools on this site, by job

Every tool below processes the file in the browser. The third column is what stays on your device.

JobToolWhat stays local
Compress JPG, PNG or WebP, or hit a target sizeImage compressorThe image
Resize or crop to a presetResize imageThe image
Convert between PNG, JPG and WebPImage format converterThe image
Images to PDF, PDF to images, merge, splitDocFlow PDF toolsThe images and PDFs
App icon at 512 × 512 for Google PlayGoogle Play icon resizerThe icon
Privacy policy for an appPrivacy policy generatorYour answers and the policy
Sitemap from a list of URLsSitemap generatorThe URL list
QR codeQR code generatorThe text or link
PasswordPassword generatorThe settings and the result
Word and character countWord counterThe text
HEX, RGB and HSL conversionColour converterThe colour values
The image compressor on this site: quality slider, output format, target size and resize fields, with the EXIF metadata removal switch on by default.
The image compressor on this site: quality slider, output format, target size and resize fields, with the EXIF metadata removal switch on by default.

The full list is on the tools page.

When a local tool is the better choice

  • The file is personal or confidential.
  • You are on a slow or metered connection.
  • You need the tool to keep working offline.
  • You do not want another account for a two-minute job.

When it is not enough

Some work is too heavy or too specialised for a browser tab: long video transcoding, some PDF operations, and some kinds of OCR need more processing power or larger models than a page can carry. For those, choose a tool that is explicit about how it handles your data and for how long. The rules for choosing one are in are online PDF converters safe. For Arabic and English document scanning in particular, the DocFlow Scanner app runs OCR on the phone rather than in the cloud.

"Local" does not mean "nothing can go wrong"

A browser tool keeps the file off other people's servers. It does not protect the file on your own machine: the download sits in your Downloads folder, a shared computer keeps it, and a synced browser profile may carry history elsewhere. Private browsing reduces the local traces; it changes nothing about whether a service uploads.

And marketing words are not evidence. "Secure", "private", "encrypted" and a padlock in the address bar describe the connection or the intent, not where the processing happens. The Network tab does.