
Make images from a prompt or an image reference, and pick the model that fits.

| Input | Text and images |
|---|---|
| Image references | Up to 10 |
| Aspect ratios | 2.4:1, 21:9, 2:1, 1.85:1, 16:9, 3:2, 4:3, 5:4, 6:5, 1:1, 5:6, 4:5, 3:4, 2:3, 9:16, 1:2 |
| Maximum resolutionSizes: Small, Medium and Large. | 2048 × 2048 at 1:1Sizes: Small, Medium and Large. |
| Quality | Fast, Quality |
| Images per generation | 1 to 6 |
| Transparent background | Not supported |
| Prompt length | Up to 9,999 characters |
On Leonardo.Ai, you use Seedream 5.0 Pro, ByteDance's image model, to make up to 6 images at a time from a prompt and up to 10 image references, with the words you write set in the image.
Yes. Open your image on Leonardo.Ai, mark the area to change with Annotate, and say what goes there.
Seedream 5.0 Pro's token cost depends on the image size and how many images you make, with the largest sizes costing more. The generation panel shows the exact cost before you generate.
Yes. You can call Seedream 5.0 Pro from the Leonardo.Ai API. The Seedream 5.0 Pro API guide lists the request fields.
On Leonardo.Ai, Seedream 5.0 Pro lets you choose Fast or Quality and set a fixed seed. Seedream 5.0 Flash is a separate model that you pick in the generator.
With Seedream 5.0 Pro on Leonardo.Ai, you can mark changes on an image with Annotate, add up to 10 image references, and choose Fast or Quality.
