Edit or combine up to 10 images with a text instruction using Meta Muse Image Edit. Try free — pay per generation, no subscription. Meta Muse Image Edit applies a text instruction to up to 10 reference images at once, for instruction-driven edits, composites, and style transforms.
Meta Muse Image Edit applies a text instruction to up to 10 reference images at once, for instruction-driven edits, composites, and style transforms.
About this model
Meta Muse Image Edit applies a text instruction to up to 10 reference images at once, for instruction-driven edits, composites, and style transforms. Describe the change you want — swap an element, match lighting across images, or combine multiple references into one scene — and Muse Image handles the rest. Use Muse Image Text to Image first to generate a base image, or compare against FLUX Kontext and Qwen Image Edit for multi-image editing.
Cost analysis
| Provider | Cost | Notes |
|---|---|---|
| muapiapp | $0.02/run (base) | Pay-per-generation pricing, no subscription required. Final charge scales with output size and quality. |
| Fal.ai | Not available | Meta Muse Image is not yet listed on Fal.ai. |
| Replicate | Not available | Meta Muse Image is not yet listed on Replicate. |
Pay-per-generation pricing, no subscription required. Final charge scales with output size and quality.
Meta Muse Image is not yet listed on Fal.ai.
Meta Muse Image is not yet listed on Replicate.
** Competitor pricing is estimated based on similar model architectures and usage tiers.
Configuration schema
| Parameter | Type | Description | Default |
|---|---|---|---|
| Prompt | string | Text instruction describing the desired edit or transformation. | Combine these into one scene, matching the lighting and color grade of the first image. |
| Image URLs | array | Reference images to edit or combine. Supports up to 10 images. | https://d3adwkbyhxyrtq.cloudfront.net/ai-images/186/712345784292/4a8c5c70-abcc-4920-873e-b0e219986453.jpg |
| Aspect Ratio | Enum (6 options) | Output layout dimensions. Leave empty to use the source image's dimensions. | - |
| Output Format | Enum (3 options) | Image file format. | webp |
Text instruction describing the desired edit or transformation.
Combine these into one scene, matching the lighting and color grade of the first image.Reference images to edit or combine. Supports up to 10 images.
https://d3adwkbyhxyrtq.cloudfront.net/ai-images/186/712345784292/4a8c5c70-abcc-4920-873e-b0e219986453.jpgOutput layout dimensions. Leave empty to use the source image's dimensions.
-Image file format.
webpDeveloper documentation
Upload 1–10 reference images via images_list.
Write an edit instruction describing the desired transformation, composite, or style change.
Optionally set aspect_ratio — leave empty to use the source image's dimensions.
Choose output_format — webp (default), png, or jpeg.
Submit the request:
curl -X POST https://api.muapi.ai/api/v1/muse-image-edit \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt": "Combine these into one scene, matching the lighting and color grade of the first image", "images_list": ["https://example.com/a.jpg", "https://example.com/b.jpg"], "output_format": "webp"}'
/api/v1/predictions/{request_id}/result until completed.Frequently asked
It applies a text instruction to up to 10 reference images at once, for instruction-driven edits, composites, and style transforms.
Between 1 and 10 images via the `images_list` field.
A base flat $0.02 per run; the final charge scales with parameters like output size and quality.
webp (default, compact web-friendly output), png, and jpeg, set via `output_format`.
Yes — use [Muse Image Text to Image](/playground/muse-image-text-to-image) to generate a starting image, then edit it (or combine it with other references) here.