Meta

This page is auto-generated from model configurations. Last updated: 2026-09-01.

This reference lists all available Meta image generation models and their parameters. Use these parameter names when calling the Generation API.

• Meta Muse Image
• Meta Muse Image Edit
• SAM 3.1 Image

---

Meta Muse Image

Generate high-fidelity images with faithful instruction-following — Meta Muse Image renders text, plots, and fine details accurately.

Model ID: model_meta-muse-image

Capabilities: txt2img

LLM Markdown: https://app.scenario.com/api/models/model_meta-muse-image/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
prompt	string	Yes	-	-	-	-	Describe the image you want to generate, including subject, style, mood, composition, and any text that should appear in the image.
aspectRatio	string	No	1:1	-	-	21:9, 16:9, 4:3, 3:2, 1:1, 2:3, 3:4, 9:16, 9:21	The aspect ratio of the generated image.
numOutputs	number	No	1	1	4	-	How many images to generate from the same prompt. Useful for exploring variations. Each additional image adds to the cost.

Meta Muse Image Edit

Edit images with precise, instruction-faithful changes and compose from multiple reference images — powered by Meta Muse Image.

Model ID: model_meta-muse-image-edit

Capabilities: img2img

LLM Markdown: https://app.scenario.com/api/models/model_meta-muse-image-edit/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
referenceImages	file_array	Yes	-	-	-	-	Images to edit or compose from. Provide 1 to 10 photos, illustrations, or designs — the model will modify them based on your instructions.
prompt	string	Yes	-	-	-	-	Describe how the input image(s) should be edited, including visual changes, style, lighting, composition, and any text that should appear.
aspectRatio	string	No	1:1	-	-	21:9, 16:9, 4:3, 3:2, 1:1, 2:3, 3:4, 9:16, 9:21	The aspect ratio of the edited image.
numOutputs	number	No	1	1	4	-	How many edited versions to generate from the same prompt and source image(s). Useful for exploring variations. Each additional image adds to the cost.

SAM 3.1 Image

Segment images with Meta SAM 3.1: text and/or box prompting, or interactive tracking (points, mask).

Model ID: model_meta-sam-3-1-image

Capabilities: img2img

LLM Markdown: https://app.scenario.com/api/models/model_meta-sam-3-1-image/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
image	file	Yes	-	-	-	-	Image to segment
imagePromptingText	string	No	-	-	-	-	Text prompt for prompt-based segmentation. Provide this and/or bounding boxes under Image prompt boxes. Omit when using only tracking (points or mask). Use a short noun phrase (e.g. ‘red car’), not a full sentence: prompts are capped at 120 characters (about 20 words).
imagePromptingBoxes	inputs_array	No	-	-	-	-	Bounding boxes with labels for prompt-based segmentation. Omit entirely when using only text or tracking. When you add a row, set all four bounds (x min/max, y min/max); partial rows may be rejected by the segment service.
imageTrackingPoints	inputs_array	No	-	-	-	-	Points for interactive tracking-based segmentation: segments the exact region under the points rather than searching for lookalikes. Label 1 = include this point, 0 = exclude it. Omit entirely when using text or box prompting. When you add a row, set both X and Y; partial rows may be rejected by the segment service.
imageTrackingMask	file	No	-	-	-	-	Existing mask to refine with tracking. Combine with tracking points to correct a previous segmentation. Omit when using text or box prompting.
applyMask	boolean	No	false	-	-	-	When enabled, returns the source image with the mask applied (PNG alpha channel, transparent background) instead of binary mask outputs.
seed	number	No	-	0	-	-	Random seed for reproducible segmentation. Leave blank to randomize.