Stability AI Inpainting
Use segmentation masks to inpaint objects within an image.
The block wraps Stability AI inpainting API and let users use instance segmentation results to change the content of images in a creative way.
Type identifier
Use the following identifier in step "type" field: roboflow_core/stability_ai_inpainting@v1 to add the block as a step in your workflow.
Properties
Name
Type
Description
Refs
name
str
Enter a unique identifier for this step..
❌
prompt
str
Prompt to inpainting model (what you wish to see)..
✅
negative_prompt
str
Negative prompt to inpainting model (what you do not wish to see)..
✅
api_key
str
Your Stability AI API key..
✅
invert_segmentation_mask
bool
Invert segmentation mask to inpaint background instead of foreground..
✅
preset
StabilityAIPresets
Optional preset to apply when outpainting the image (what you wish to see). If not provided, the image will be outpainted without any preset. Avaliable presets: 3d-model, analog-film, anime, cinematic, comic-book, digital-art, enhance, fantasy-art, isometric, line-art, low-poly, modeling-compound, neon-punk, origami, photographic, pixel-art, tile-texture.
❌
seed
int
A specific value that is used to guide the 'randomness' of the generation. If not provided, a random seed is used. Must be a number between 0 and 4294967294.
✅
The Refs column marks possibility to parametrise the property with dynamic values available in workflow runtime. See Bindings for more info.
Runtime compatibility
requires_internet - air-gapped / offline deployments : This block depends on a service that is not reachable from fully offline / air-gapped deployments.
Input and Output Bindings
The available connections depend on its binding kinds. Check what binding kinds Stability AI Inpainting in version v1 has.
Input and output bindings
input
image(image): The image to inpaint..segmentation_mask(instance_segmentation_prediction): Model predictions from segmentation model..prompt(string): Prompt to inpainting model (what you wish to see)..negative_prompt(string): Negative prompt to inpainting model (what you do not wish to see)..invert_segmentation_mask(boolean): Invert segmentation mask to inpaint background instead of foreground..seed(integer): A specific value that is used to guide the 'randomness' of the generation. If not provided, a random seed is used. Must be a number between 0 and 4294967294.
output
image(image): Image in workflows.
Last updated
Was this helpful?