The previous version of the GPT Image line: it draws images from a description and edits your photos. Three sizes and three quality levels, from two cents. A good fit when you need an inexpensive image and no large sizes.
Create an image with GPT Image 1.5

| What it does | Image from a description and photo editing |
|---|---|
| Sizes | 1:1, 3:2, 2:3 |
| Quality | Low, medium, high |
| Price per image | $0.02 to $0.26 |
| Photo edit price | $0.02 to $0.26 |
| Photos as input in edit | up to 3 |
Square 1:1, landscape 3:2 and portrait 2:3. That covers posts, stories and covers, and there are no extra options in the list.
Low costs two cents and is enough to test an idea. Medium and high give more detail and cost noticeably more.
The same model changes the background, clothes and details on an uploaded photo. In editing you can keep the proportions of the source.
Editing takes up to three source photos at once, for example a person and an object to combine into one frame.
The cost depends on size and quality and is shown before you run. If a generation fails, the money is returned.
Version 1.5 has three sizes and no 4K. The newer GPT Image 2 holds the description more closely, writes cleaner text and supports sizes up to 3840×2160.
Take low to check an idea: it costs two cents. For a finished creative, medium or high is better.
Yes, the “Edit” tab works with this model. You can upload up to three photos and keep the proportions of the source.
$0.02 to $0.26. The price depends on quality and size, and you see it before you run.
No. You pay for each image at the price shown before you run.