silhouette-2024.5 : Nodes : Stability
Stability
Description
Stable Diffusion is a pioneering, deep learning text-to-image model. It is primarily used to generate detailed images based on text descriptions. The Stability node integrates Stable Diffusion models directly in Silhouette. Inpaint, outpaint or generate images from scratch using text prompts.
Note: Internet access, a DreamStudio account and agreement to Stability AI’s Terms of Service are required to use Stability. Sign up, go to Your Account > API Key, copy the key and paste it into the Machine Learning > Stability AI API Key preference.
Node Group
Image.
Go to the Stability tutorial to see how it works.
Controls
Mode
Generate
Generates an image based on a text prompt. An input image is not required, but when present, acts as an additional prompt.
Inpaint
Creates an image inside the alpha.
Outpaint
Creates an image outside the alpha.
Model
Choose the Stability model version.
Prompt
Enter a text prompt to produce an image. For more information on prompts, see the Prompt Guide.
Credits
The amount of remaining credits in your DreamStudio account are displayed once you press Generate.
Generate
Generates the images from text, inpaints or outpaints.
Composite
Composites the result using the source alpha and is useful if you need to blur the alpha’s edge. This is only available in the Inpaint and Outpaint modes.
Options
Max Size
Auto
Automatically sets the Max Size based on the session size. For sessions larger than 1024 pixels, the image is scaled down to the to 1024 limit and the result is scaled back up to the session size.
512
The size is set to 512 x 512 pixels.
768
The size is set to 768 x 768 pixels.
1024
The size is set to 1024 x 1024 pixels. Currently, this is the maximum size.
Prompt Strength
How much the prompt influences the result.
Image Strength
If an input image is present, determines how much the image influences the result.
Steps
The more steps, the higher the quality. Lower steps generate more quickly and cost less, but if they are lowered too much, image quality will suffer. Higher steps cost more and take longer to generate, but produce more detailed results. The improved quality becomes less drastic as the step value increases.
Seed
Iterates and randomizes the existing result. Using the same seed produces the same result.
Number of Images
How many images to generate.
Images
Displays the generated images.
Delete
Select one or more images, right-click and choose Delete.