Give a requirement — size, dimensions, format — and get an image that meets it.
Or choose a file, or paste a screenshot. It is measured and re-encoded on your device and never uploaded.
Fitting on your device…
Your image is opened, measured and re-encoded in this browser. It is never uploaded, and nothing about it is stored.
Most image tools ask you to be the expert. They offer a quality slider and a width box and leave you to work out which combination the form in front of you will accept. This one works the other way round: you give it the requirement, and it finds the settings.
Type a maximum file size, or exact dimensions, or both. The tool encodes at the highest quality that fits, and shows you a line-by-line report of what your file is against what was asked for. If the requirement cannot be met, it tells you the closest achievable size instead of handing you something that will be rejected.
Portals rarely say how far over the limit you are. Put the number they quoted into the size box and you get a file that is under it, at the best quality that fits.
"1200 × 1200, JPEG, under 2 MB" normally means three tools and three guesses. Here it is one pass, and the report confirms all three.
Drop in an image that was rejected and read the report. It tells you which requirement it actually failed, which is usually not the one you assumed.
The named target pages cover common requirements. This is where the uncommon ones go — an internal system, a client's brief, a printer's spec sheet.
Drag a file onto the page, choose one, or paste a screenshot. Nothing is uploaded — the work happens on your device.
Every requirement is listed with a tick or a cross against what your file actually is. Anything the rule asks for that software cannot check is listed too, unticked.
The file you get is the one that passed. If nothing could pass, you are told the closest achievable size instead of being handed something that will be rejected.
No. It is decoded, drawn to a canvas and re-encoded entirely in your browser. Nothing is sent to a server, so nothing can be stored, logged or seen.
You are told so, with the number: "the smallest this can be at 600 × 600 is 248 KB, and the limit is 240 KB". Other tools either destroy the image trying or quietly hand back something over the limit. A photo with a plainer background will usually get there — fine detail is what costs the bytes.
Because a squashed face is a rejected photo. When you give exact dimensions the default is to fill the frame and crop the overflow. Choose "Pad to fit" if the whole image has to stay visible, which is usually right for product shots on white.
Yes, and it says so in the report. The pixels are interpolated rather than real detail, so the file meets the rule without actually being sharper. That is worth knowing before you submit it.
Yes, as a side effect of re-encoding — including any GPS coordinates your camera wrote. If that is all you want, the EXIF Remover does it without touching the pixels.