Convert an image to banner size
Fit a whole photo into a LinkedIn, X or YouTube banner without cutting anything off.
A landscape photo is the easiest start; a portrait one works with more built.
- Free
- No sign-up
- Photos stay on your device
LinkedIn 1584 × 396, X 1500 × 500, a 1200 × 630 link card or YouTube channel art: the whole photo fitted in, the long sides built, and the profile-picture overlap marked.
Opens at 1584 × 396 with mirrored edges and the profile-picture overlap shaded. The chips switch banner; safe areas are drawn on the canvas, never in the file.
A 4000 × 3000 photo in each banner
| Banner | Photo drawn at | Built | Model |
|---|---|---|---|
| LinkedIn background, 1,584 × 396 | 528 × 396 (13.2%) | 67% | not offered |
| X header, 1,500 × 500 | 667 × 500 (16.7%) | 56% | not offered |
| Link card, 1,200 × 630 | 840 × 630 (21%) | 30% | offered, 4 passes |
| YouTube channel art, 2,048 × 1,152 | 1,536 × 1,152 (38.4%) | 25% | offered, 6 passes |
Why very wide shapes are different
A 4:1 banner from a 4:3 photo leaves two strips beside the picture that are each wider than half the photo. That is beyond what any continuation of the edge can carry convincingly, so this page leans on the fills whose failure you can see coming: a mirror, which repeats the real edge, and smart extend, which runs the edge's colours outward and softens them. Look at the joins at 200% in the Fill tab before you save.
The awkward questions
How do I fit a whole photo into a 4:1 LinkedIn banner?
Pick the LinkedIn chip. The photo is scaled to fit the banner's height and placed whole; the long sides are built — mirrored to start, with smart extend one tap away. Nothing is cut off the top or the bottom, which is what a crop to 4:1 would do to almost any photo.
Why is the AI fill not offered on banners?
At 4:1 the new region is far larger than the photo beside it. This canvas is 3.0 times the area of your photograph, and the model is only offered up to 2. Past that, more than half of what you saved would be generated rather than photographed, and a continuation larger than the thing it continues stops being a continuation. The seven deterministic fills have no such limit. Mirror and smart extend are pre-selected because they are honest at that width.
What are the shaded boxes on the canvas?
On 4:1 and 3:1 banners, the lower-left area a profile picture covers on a desktop; on 16:9 channel art, the centre band that shows on every device. They are guides we observed, labelled “unverified” — platforms move these — and they are never saved into the file.
Are these the official sizes?
They are the sizes in common use, carried from our size list and labelled as not yet re-verified with a dated source. The custom fields take any other width and height, and the exact numbers are printed on the Save button.
Is my photo uploaded?
No. It is fitted, filled and saved by this tab.
Related
- YouTube thumbnail sizeAny picture into a 16:9 thumbnail at 1280 x 720 or larger, fitted whole, with the duration-badge corner marked.
- Extend an image on one sideWiden one side, or two, to make room for a title or a crop-safe margin — the rest of the frame stays put.
- Smart extendThe seam's texture carried across and its colours run outward along the lines meeting the frame. No model, same every time.
- Aspect ratios and sizesEvery common ratio and platform size, with the canvas two real photos would need to reach it and how much of it is built.