Image Resizer

No. Decoding and re-encoding happen entirely in your browser using the Canvas API. Your image never leaves your device, which makes this safe for private or sensitive pictures.

Pixels sets an exact width and height. Percent scales by a factor of the original, so 50 percent halves each side. Fit box shrinks the image to sit inside a maximum width and height while keeping its shape. Longest edge makes the longer side a set number of pixels and scales the other side to match.

Making an image smaller keeps it sharp, and the canvas uses high-quality smoothing. Enlarging past the original size cannot add real detail, so upscaling is off by default in Fit box and Longest edge and must be enabled with the Allow enlarging checkbox.

PNG (lossless, keeps transparency), JPEG (small, best for photos, filled on a white background), and WebP (modern, usually the smallest). A quality slider from 10 to 100 percent appears for the lossy formats JPEG and WebP.

Yes by default, so images never look stretched. In Pixels mode you can untick Lock aspect ratio to force an exact width and height; Fit box and Longest edge always preserve the shape.

Re-encoding a heavily compressed JPEG as lossless PNG, or raising the quality slider, can increase the byte size even at the same dimensions. The readout shows the change as a percentage so you can spot this; switch to WebP or lower the quality to shrink it.

In Fit box mode the presets fill the max width and height for common targets: 4K, 1080p, 720p, Instagram square and portrait, a Twitter or X header, a thumbnail and an avatar. They are a starting point you can still edit by hand.

Only your device's memory. Very large images use more memory to decode and draw, but everything stays local with no upload cap, and the preview updates a moment after each change.

Yes. Because the image is redrawn to a fresh canvas and re-encoded, the original EXIF metadata, including any GPS coordinates, is not carried into the downloaded file.