Basic Information

appmixer.ai.huggingface.GenerateImage

Generate Image

1.0.0

Public

Generate an image from a text prompt using a text-to-image model served by Hugging Face Inference Providers. The image is saved to file storage and its file ID is returned.

Input Ports

in

FieldTypeDescription
Model text Model path as the selected provider names it, e.g. fal-ai/flux/schnell for fal.ai. This is the provider path, not always the Hub repository ID.
Provider select Inference Provider that serves the model. The legacy hf-inference provider no longer serves text-to-image and answers HTTP 410 for most checkpoints, so fal-ai is the default.
Prompt textarea Description of the image to generate, e.g. a watercolour painting of a lighthouse at dawn.
Negative Prompt textarea What the image should avoid. Not supported by every model.
Width number Width of the generated image in pixels. Leave empty for the model default.
Height number Height of the generated image in pixels. Leave empty for the model default.
Inference Steps number How many denoising steps to run. More steps usually means better quality and a slower, more expensive call.
Guidance Scale number How closely the image should follow the prompt. Higher values follow the prompt more literally.
File Name text Name to store the image under. Leave empty to generate one automatically.

Output Ports

out

FieldTypeDescription
File ID string -
File Name string -
Content Type string -
Size (bytes) integer -
Model string -
Provider string -

Full Metadata