About PixelSquish

Last updated 2026-10-09 · Launched 14 August 2026

PixelSquish is a set of free web tools for compressing, resizing and converting images. One thing separates it from most alternatives: your files never travel anywhere. Everything happens inside the browser tab you are looking at right now.

Why it was built this way

The standard design for an online image tool is upload → process → download. Your file travels to a company's server, gets converted there, and comes back. There is nothing sinister about that — servers are fast and it works on any device. But it does mean that, for some window of time, a copy of your image exists on a machine you don't control.

That mattered to me when the file was a scanned contract, a photo of an ID, or product shots belonging to an employer. So this was built the other way. Browsers ship with an API called Canvas that can decode and re-encode images on their own. PixelSquish is built on it.

You can verify the claim yourself

Any site can say it doesn't upload your files. Two checks settle it, and neither needs technical skill:

• The offline test. Open a tool page, switch off Wi-Fi, then drop an image in. If it still converts, your file plainly isn't travelling anywhere — there is no connection to travel over.
• The network tab. Press F12, open the Network tab, and watch while you convert. An upload-based tool sends your file as a large POST request. This one doesn't.

Run those checks and you will still see small requests: analytics events, an anonymous usage counter and, on tool pages, the Google AdSense ad script. The analytics carry actions such as which tool was opened — never the image, and never the file name. The first time you add a PDF, HEIC or TIFF you will also see the decoder being downloaded; that is code coming to you, not your file going out. The privacy policy spells out exactly what is in them.

What it can't do

Browser-side processing is not magically better at everything, and it's fair to know the costs.

• Your device does the work. A hundred large photos on an old laptop will take as long as that laptop takes. There is no server farm behind it.
• The browser sets the limits. Formats a browser can't open on its own — iPhone HEIC, PDF and TIFF — need a decoder, so the first time you add one, an open-source decoder is downloaded and runs on your device. That makes the first conversion slower, and files the decoder can't handle (a password-protected PDF, for example) can't be opened. Upload-based tools decode on their servers and can sometimes cope with more.
• Compression is not guaranteed. A photo that was already well optimised will not shrink further. When that happens you get your original file back and the result says so — a compressor that hands you a bigger file has failed at its one job.

Who runs it

One developer builds and runs this. Not a company, not a team. I work in e-commerce operations, where resizing images and matching platform specs is a daily chore, and I built these tools for myself first. I still use them in my own work. That is usually why a feature exists — because the lack of it annoyed me.

What broke while building it, and why, gets written up on the blog. Writing only about what went well would make it advertising rather than a record, so the mistakes go in too.

How it makes money

Advertising. The site shows ads served by Google AdSense, and that is what pays for it. Every tool stays free, with no account and no payment, and no feature is locked behind the ads. Running costs are the domain ($10.46 a year) and a server that stays inside a free tier — because your device does the image work, there is almost nothing to pay for.

What the advertising cookies do, and how to turn off personalised ads, is explained in the privacy policy. Your images are never sold or shared under any circumstances — they are never received in the first place, so there is nothing to pass on.

Get in touch

Found a bug, need a tool that isn't here, or have a question? Send it through the contact page. I read them myself.