StableDiffusion3API (🚫Deprecated)(stableDiffusion3API)

The StableDiffusion3API node is a deprecated API-based Stable Diffusion 3 image generation node from the ComfyUI-Easy-Use package. It accepts a positive prompt, a negative prompt, model and aspect ratio selections, a seed, a denoise amount, and an optional input image, then returns a generated image. The node appears under EasyUse/🚫 Deprecated and is kept for backwards compatibility.

Inputs

• positive (STRING, required): The positive text prompt describing what should appear in the generated image.
• negative (STRING, required): The negative text prompt describing elements to avoid or exclude from the generated image.
• model (COMBO, required): The Stable Diffusion 3 model to use. Available options are sd3 and sd3-turbo. sd3-turbo is aimed at faster generation, while sd3 is the standard model.
• aspect_ratio (COMBO, required): The aspect ratio of the output image. Available options are: 16:9, 1:1, 21:9, 2:3, 3:2, 4:5, 5:4, 9:16, and 9:21.
• seed (INT, required, default=0): The random seed used for generation. Values range from 0 to 4294967294. Using the same seed with the same inputs produces repeatable results.
• denoise (FLOAT, required, default=1): The amount of denoising applied during generation. Minimum is 0, maximum is 1. Lower values preserve more of the optional input image, while higher values allow the prompt to drive a stronger change.
• optional_image (IMAGE, optional): An optional input image used as a starting point for image-to-image generation. When connected, the output is influenced by this image based on the denoise value.

Outputs

• IMAGE: The generated image as a ComfyUI image tensor.

Usage Notes

Because this node is deprecated, prefer any newer or replacement Stable Diffusion 3 API nodes provided by the package when building new workflows. This node is primarily useful for maintaining compatibility with existing graphs.

For text-to-image generation, provide only the prompt inputs and leave optional_image disconnected. For image-to-image workflows, connect an image to optional_image and adjust denoise to control how much of the original image is retained. A denoise value of 1.0 allows the strongest transformation, while lower values keep the output closer to the source image.