GrokImageNode - ComfyUI Built-in Node Documentation

The Grok Image node generates one or more images based on a text description using the Grok AI model. It sends your prompt to an external service and returns the generated images as tensors that can be used in your workflow.

Inputs

Parameter	Description	Data Type	Required	Range
`model`	The specific Grok model to use for image generation. Different models may offer varying quality, speed, or features.	COMBO	Yes	`"grok-imagine-image-quality"` `"grok-imagine-image-pro"` `"grok-imagine-image"` `"grok-imagine-image-beta"`
`prompt`	The text prompt used to generate the image. This description guides the AI on what to create. Must be at least 1 character long.	STRING	Yes	N/A
`aspect_ratio`	The desired width-to-height ratio for the generated image.	COMBO	Yes	`"1:1"` `"2:3"` `"3:2"` `"3:4"` `"4:3"` `"9:16"` `"16:9"` `"9:19.5"` `"19.5:9"` `"9:20"` `"20:9"` `"1:2"` `"2:1"`
`number_of_images`	Number of images to generate (default: 1).	INT	No	1 to 10
`seed`	A seed value to determine if the node should re-run. The actual image results are nondeterministic and will vary even with the same seed (default: 0).	INT	No	0 to 2147483647
`resolution`	The desired output resolution for the generated images (default: “1K”).	COMBO	No	`"1K"` `"2K"`

Note: The seed parameter is primarily used to control when the node re-executes within a workflow. Due to the nature of the external AI service, the generated images will not be reproducible or identical across runs, even with an identical seed. Note on pricing: The cost of generating images depends on the selected model, resolution, and number_of_images. For example, the “grok-imagine-image-quality” model with “1K” resolution costs

0.05 per image, while "2K" resolution costs

0.07 per image. The “grok-imagine-image-pro” model costs

0.07 per image, and other models cost

0.02 per image.

Outputs

Output Name	Description	Data Type
`output`	The generated image or a batch of images. If `number_of_images` is 1, a single image tensor is returned. If greater than 1, a batch of image tensors is returned.	IMAGE

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Source fingerprint (SHA-256): f4af858c6b94fd74be1856849417609f3d45fc0f139e8f2e56cdabb6e3521076

​Inputs

​Outputs

Inputs

Outputs