Vision: Image to Image

Using a text prompt and seed image to generate an image.

Log in to see full request history
Body Params
string
required

Base64 encoded seed image.

text_prompts
array of objects

Prompts for image generation.

text_prompts
float
Defaults to 0.25

The weight of the initial image. Accepts 0-0.9

int32
Defaults to 2

Classifier-Free Guidance Scale: Higher values follow text input rigorously. Accepts 1-10

int32
Defaults to 8

Number of steps in the image generation process. Proteus/Dreamshaper/Stable Diffusion: Accepts 6-12 Playground: Accepts 25-51.

string
Defaults to proteus
string
Defaults to 1024
string
Defaults to 1024
Responses

Language
Credentials
Click Try It! to start a request and see the response here! Or choose an example:
application/json