MetaMuseImageTextToImageApi - ComfyUI Built-in Node Documentation - ComfyUI

Meta Muse Image Text to Image generates images from a text prompt using Meta’s Muse Image model. The model reasons about the prompt before rendering, and can use web search, image search, and code execution while planning the image. The node calls the Muse Image API and returns the resulting image or images.

Inputs

Common Inputs

Parameter	Description	Data Type	Required	Range
`model`	Model to use.	DYNAMIC_COMBO	Yes	`"muse-image-1.0"`

Selecting a model in the list shows the settings that model supports. The only available model is muse-image-1.0; its settings are listed below.

muse-image-1.0 Inputs

Parameter	Description	Data Type	Required	Range
`prompt`	Prompt describing the image. The model reasons about the prompt, and may use its built-in web and image search, before rendering.	STRING	Yes	Multiline text, minimum 1 character
`aspect_ratio`	Aspect ratio of the output. Images are rendered at about 2.5 megapixels (1:1 is 1600x1600, 16:9 is 2048x1152); “auto” lets the model choose from the prompt.	COMBO	Yes	`"auto"` `"1:1"` `"3:2"` `"2:3"` `"4:3"` `"3:4"` `"5:4"` `"4:5"` `"16:9"` `"9:16"` `"21:9"` `"9:21"` `"2:1"` `"1:2"`
`reasoning_strength`	How much the model thinks, plans and self-refines before rendering.	COMBO	Yes	`"high"` `"low"`
`enable_web_search`	Lets the model search the web for facts and live information while planning the image.	BOOLEAN	No	True False (default: True)
`enable_image_search`	Lets the model search for reference images while planning the image.	BOOLEAN	No	True False (default: True)
`enable_shell`	Lets the model run code while planning, for precise layouts, charts and diagrams; when off, quantities and alignment are approximated.	BOOLEAN	No	True False (default: True)
`seed`	Seed to determine if node should re-run; the API has no seed, so actual results are nondeterministic regardless of this value.	INT	Yes	0 – 2147483647 (default: 42)

Note: The prompt must contain at least one character. When aspect_ratio is set to “auto”, no explicit size is sent to the API and the model decides the output size from the prompt. The seed parameter only controls when the node re-runs; it is not sent to the API, so the generated results are nondeterministic.

Outputs

Output Name	Description	Data Type
`image`	The generated image returned by the API, decoded and provided as a batched image. If the API response contains multiple images, they are combined into one batch.	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): 59ebd72fab3db44a35ceac723606de4eabb5fe2b690d0b701db50e0e22a9e699