Resize Images for Instagram

Instagram will accept almost any photo you give it, but it does not keep what you gave it: anything that does not match one of its canvases gets cropped or padded on the platform’s terms, centred, without asking. Preparing the pixels yourself is how you keep the framing decision. This page resizes images for Instagram at the three dimensions the platform actually uses — square, portrait and story — with Contain fitting so no pixel of your photo is cut away. Drop in up to five pictures, click a preset, and download files that land on the feed looking exactly as composed. The resizing runs entirely in your browser; your photos are never uploaded.

Drop images here, click to browse, or paste from clipboard (Ctrl+V)

JPEG, PNG or WebP · up to 30 MB per file · up to 5 images at a time · processed locally

The three canvases worth preparing for

Feed posts render at 1080 pixels wide, and two shapes matter. The square canvas is 1080 × 1080, the classic 1:1 frame that fits every feed layout ever shipped. The portrait canvas is 1080 × 1350 — a 4:5 ratio, and the tallest a feed post may occupy — and it matters more than its obscurity suggests: a portrait post covers roughly a quarter more screen than a square one, which is why accounts that test both rarely go back. Stories and Reels live on 1080 × 1920, a 9:16 full-screen canvas where vertical content was always the point.

Why 1080? It is the display resolution the platform standardised on, and uploads larger than that are simply downscaled on arrival — harmless, but wasted bytes on the way in, and a missed chance to control the resampling yourself. Preparing files at these exact dimensions means what you composed is what appears, with no platform-side interpretation of where the crop should fall.

Downscale freely, upscale carefully

Nearly every phone today shoots at 3000 to 4000 pixels on the long edge, which makes preparing for Instagram a downscaling job — the safe direction. Shrinking uses the surplus data a modern camera captures to produce a crisp, dense 1080-wide image; nothing is invented, detail is only consolidated. The result cards show original and output dimensions side by side so the change is explicit.

The opposite direction is where honesty is owed. Enlarging — a 900-pixel image asked to become 1080, or a screenshot stretched to story height — cannot create detail the source never had; the encoder interpolates between existing pixels, and the output is softer than a native capture at that size. The page shows a warning whenever a target exceeds the source rather than pretending otherwise. Small enlargements are usually invisible; large ones are not, and if the preview looks soft, the honest fixes are a smaller canvas or a better source file.

What Contain does when a photo does not match

The three canvases have fixed proportions, and a photo with a different shape has to meet them somehow. This page fixes the fitting rule to Contain: the entire image is scaled to fit inside the canvas, and any leftover space is padded — white where the kept format is JPEG, transparent for PNG and WebP sources. Nothing is cropped, ever; a wide landscape becomes a square with bands above and below, its composition intact.

Contain is the right default when composition is not negotiable — documents, products, screenshots, anything where cutting an edge changes the meaning. When the destination is a photograph and full-bleed looks better, the platform-style alternative is Cover, which crops the overflow from the centre; that mode, along with stretch and the by-width and by-height options, lives on the full resizer page, and the two tools compose: resize to the canvas here, and if a byte limit also applies, finish with the target-size compressor.

Frequently asked questions

Which canvas should my post use?

Match the photo’s native shape first: portrait photographs belong on 1080 × 1350, where they earn the most feed space; genuinely square compositions are happiest at 1080 × 1080; and anything shot vertically for full-screen viewing — plus every Story and Reel — wants 1080 × 1920. If a single image must serve both feed and story, prepare both files from the original rather than stretching one output across both canvases.

What happens if I upload a photo with the wrong ratio?

The platform intervenes: feed uploads that exceed the allowed height are cropped to fit, centred, with no preview of what will be cut, and stories handle mismatched shapes by letterboxing or cropping depending on version and settings. Those rules change silently and differ by surface, which is exactly why handing the platform a pre-sized file wins — the crop decision was made by you, at your desk, where it can be undone.

Does resizing change the file size too?

As a side effect, yes — fewer pixels mean fewer bytes, so a 4000-pixel phone photo brought down to 1080 wide typically drops from several megabytes to a few hundred kilobytes even before any compression. But the number is only influenced, not controlled: if the destination also enforces a byte ceiling, resize first and then compress to the stated limit with the dedicated target-size tool, so each dial is set deliberately and only once.

Is it bad to upscale a small image to Instagram’s dimensions?

It is not forbidden, just visibly worse. Upscaling interpolates new pixels between existing ones, and a 600-pixel image stretched to 1080 will look soft next to one captured natively — the warning on this page appears the moment any target exceeds the source. The platform’s own downscaling will not rescue it. Where the choice exists, shoot or export closer to the target size; where it does not, a modest enlargement of 20 to 30% is usually imperceptible.

Should I sharpen or filter my photos before resizing?

Do all editing before the resize, and make the resize the final step before upload. Sharpening or filtering a freshly downscaled image can exaggerate the resampling into halos, and the reverse order — resize then edit — forces your editor to work on fewer pixels than the platform will ultimately display. This page deliberately does one thing; hand it finished images and its output is the last step before upload.