Social Image Resizer

Turn one picture into every crop you need — square, portrait, story and landscape — with a safe-area guide and repositioning that works with a mouse, a finger or buttons.

Why one image never fits every platform

A photograph has one aspect ratio. Social platforms have many, and they crop to their own without asking. A landscape shot posted to a full-screen story gets the sides thrown away; a portrait shot in a link preview gets the top and bottom cut off. The part you cared about is often the part that disappears.

The fix is not to resize, it is to re-crop. Resizing squashes or letterboxes; cropping chooses which part of the frame survives. This tool does the second, and shows you the result for each target size before you commit.

There are four ratios that cover most of what people need. Square at 1:1 is the safe default and is rarely cropped further. Portrait at 4:5 is the tallest most feeds will show without trimming. 9:16 is the full-screen story and reel shape. 16:9 is the landscape shape used by video thumbnails and most link previews.

Start from the largest, sharpest original you have. Every crop discards pixels, so a small image cropped to a tall shape can end up too low-resolution to look good. Cropping down from a big original is always better than scaling a small one up.

Choose one image. It is read in your browser and never uploaded.
Pick the target sizes you need, or enter custom dimensions or a custom ratio.
Reposition the image: drag it, use the arrow buttons, or type an exact percentage. All three do the same thing.
Turn on the safe-area guide to see roughly where platform buttons and captions may sit.
Choose a background for any gaps: a solid colour, or a blurred copy of the image itself.
Export as JPEG, PNG or WebP, and download one crop or all of them as a ZIP.

Safe areas, backgrounds and the things that ruin a crop

Full-screen formats are the ones that catch people out. A story or reel fills the whole phone screen, and the platform puts its own interface on top: a profile chip and a close button near the top, captions and a reply box near the bottom. Anything you place in those bands is liable to be covered.

The safe-area guide draws a dashed rectangle showing the region that is usually left alone, and dims the parts that are not. It is a rule of thumb rather than a specification, because every platform positions its controls slightly differently and moves them between releases. Keep faces, logos and text inside the rectangle and you will be fine on almost all of them.

When an image does not fill a frame you have two honest choices. A solid background is clean and predictable, and a white or black surround often looks deliberate. A blurred copy of the image itself fills the gap with something that belongs to the picture, which usually reads better for photographs; this tool over-scales the blurred layer so there is never a visible seam at the edge.

Two habits will save you most of the trouble. First, do the crop last: export from your original rather than from an already-cropped copy, because each round of cropping and re-encoding loses a little more. Second, check the actual output dimensions rather than trusting a preset blindly. Platform requirements change, so the numbers in this tool are dated and versioned on purpose.

The preset sizes here were last reviewed on 2026-09-13. If you are reading this much later, treat them as a sensible starting point and confirm anything that has to be exact against the platform's own current guidance.

Frequently asked questions

Are my images uploaded?: No. The image is decoded, cropped and re-encoded by your own browser in this tab. There is no upload, no server processing and no analytics.
Can I use this without dragging?: Yes, completely. Arrow buttons move the image by a fixed step, and numeric inputs set the horizontal and vertical position as an exact percentage. The buttons and the numbers share the same maths as the drag gesture, so all three reach identical results.
Are the preset sizes official?: No. They are our reading of publicly documented guidance on a specific date, and that date is shown next to the presets along with a version number. Platforms change dimensions without notice, so please confirm anything critical against the platform itself.
What is the safe area?: It is the part of a full-screen crop that platform interface elements are unlikely to cover. It is an approximation, not a specification, and it varies between platforms and app versions. Keep important content inside it as a precaution.
Does the guide appear in my exported file?: No. The safe-area overlay is drawn only in the on-screen preview. Exported files contain your image and your chosen background, and nothing else.
Which export format should I choose?: JPEG for photographs, PNG when you need transparency or have sharp graphics and text, WebP when you want the smallest file and the destination supports it. JPEG and WebP are lossy; PNG is not, but produces much larger files for photographs.
Why does my exported image look softer than the original?: Because cropping to a smaller frame resamples the image, and any lossy export re-compresses it. Starting from the largest original you have, and exporting once rather than repeatedly, keeps this to a minimum.
Can I crop several images at once?: Not currently. This tool takes one source image and produces multiple crops of it. Batch conversion of many separate images is what the Image Converter & Compressor is for.

What this tool will not do

Preset sizes are a dated snapshot, last reviewed on 2026-09-13, and are versioned so you can tell how old they are. They are not official and platforms change them without notice.
The safe-area guide is an approximation. Platform interface elements differ by platform, by app version and by device, so it cannot be exact.
One source image at a time. This tool produces many crops of one picture, not one crop of many pictures.
Input is limited to JPEG, PNG and WebP. HEIC, AVIF and camera RAW are not supported because browsers cannot reliably decode them.
Cropping and exporting resample and re-encode the image, so some quality is always lost relative to the original. JPEG and WebP exports are lossy at every quality setting.
Enlarging a small image to fill a large preset cannot add detail that was never captured. Start from the biggest original you have.
The blurred background uses a canvas filter that a small number of older browsers do not implement. Where it is unavailable the solid colour is used instead, rather than an unblurred copy.
Very large images are clamped to a pixel budget that your device can safely allocate, so an extremely large output may be reduced. The tool tells you when this happens.
Colour profiles are not preserved through the canvas, so images with an unusual embedded profile may shift slightly in colour.
The ZIP download stores files without compression, because the exported formats are already compressed.