Issues with text to image

Shotstack API
1

I’m having a couple of issues with the text-to-image function in the create API. First, when I use the playground, the height and width parameters are ignored and the image always ends up square 1536px.

This is my post:

{
“provider”: “shotstack”,
“options”: {
“type”: “text-to-image”,
“prompt”: “Photo realistic image of cumulus clouds from the perspective of an airplane flying just slightly above them”,
“width”: 1024,
“height”: 576
}
}

Second, when I post the same thing using the API, I get a successful response, but the image asset doesn’t seem to exist. It’s not in my list of generated assets in the UI, nor is it found when I use the get image API call.

{“data”:{“type”:“asset”,“id”:“01j5e-qjkd1-kzfen-1ah2w-mbefxc”,“attributes”:{“owner”:“v8tlx64pi6”,“provider”:“shotstack”,“type”:“text-to-image”,“status”:“queued”,“created”:“2024-08-16T23:15:16.514Z”,“updated”:“2024-08-16T23:15:16.514Z”}}}

Any thoughts would be appreciated.

-Stephan

2

Hi @vpoids,

You can specify width and height with the stable diffusion model by referencing a font and family.

For text-to-image, you can also specify the aspect ratio.

Please refer to the AI image generation section in our documentation for more details.