Models/Image APIs/Meta Muse Image API
LiveMeta · Muse Image generation and editing2 endpoints

Meta Muse Image API — Text-to-Image & Editing

The Meta Muse Image API hub combines fast text-to-image generation with instruction-based editing and multi-image compositing. Use one Muapi workflow to create a base image, refine references, and return web-friendly output formats.

Muse Image Text to ImageFast prompt-to-image generation1 models
Text to ImageLiveFeatured
Text to Image

Muse Image Text to Image

Generate low-cost images from detailed prompts covering the subject, composition, lighting, mood, and style.

6 aspect ratios · webp / png / jpeg
$0.02/image
Try Model
Muse Image EditInstruction-based multi-image editing1 models
Image EditingLive
Edit

Muse Image Edit

Edit or combine up to ten reference images with a natural-language instruction and optional output framing.

6 aspect ratios · webp / png / jpeg
$0.02/image
Try Model

What is the Meta Muse Image API?

Meta Muse Image is a compact image family on Muapi with two complementary endpoints. Text to Image generates a new image from a detailed prompt, while Edit applies an instruction to one or more reference images for transformations and composites.

Both endpoints cost $0.02 per image and support six aspect ratios plus webp, png, and jpeg output. Edit requires images_list with up to ten source image URLs; its aspect ratio is optional so the source layout can be preserved.

Muse Image API capabilities

Prompt-to-image generation

Describe the subject, scene, composition, lighting, mood, and style in one required prompt.

Instruction-based editing

Apply natural-language transformations to one or more reference images.

Multi-image compositing

Pass up to ten image URLs to combine references into a single scene.

Web-ready formats

Choose webp, png, or jpeg output for downstream product and content workflows.

Muse Image workflows

Creative concept generation

Generate fast visual directions for campaigns, storyboards, product concepts, and social content.

Reference-led composites

Combine several source images into one scene while describing the desired lighting and color grade.

Iterative image production

Generate a base asset with Text to Image, then refine it through the Edit endpoint in the same API integration.

Quick start — Meta Muse Image API code examples

Submit a Muse Image text-to-image task with your Muapi key, receive a request ID, and poll the shared predictions endpoint for the generated image URL.

1. Submit a Muse Image job

curl -X POST https://api.muapi.ai/api/v1/muse-image-text-to-image \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"A cozy reading nook by a rain-streaked window, warm lamp light, potted plants","aspect_ratio":"16:9","output_format":"webp"}'

# Response: {"request_id":"REQUEST_ID"}

The example calls Text to Image with a prompt, aspect ratio, and webp output. Switch to muse-image-edit and include images_list for an instruction-based edit.

2. Poll the result

curl https://api.muapi.ai/api/v1/predictions/REQUEST_ID/result \
  -H "x-api-key: YOUR_API_KEY"

Poll until the task completes, then read the generated image URL from the result response.

Muse Image API FAQ

Which Muse Image endpoints are available?

Muapi lists muse-image-text-to-image for prompt generation and muse-image-edit for instruction-based editing or compositing.

How many images can Muse Image Edit use?

The images_list field accepts up to ten source image URLs. The prompt describes the requested edit or composite.

Which aspect ratios and formats are supported?

Both endpoints support 1:1, 16:9, 9:16, 4:3, 3:4, and 21:9. Output can be webp, png, or jpeg.

What happens if I omit aspect_ratio for Edit?

The Edit schema allows aspect_ratio to be omitted so the output can use the source image's dimensions.

How is Muse Image priced?

Both Text to Image and Edit cost $0.02 per image.

Build with Muse Image

Connect low-cost prompt generation and multi-image editing through one Muapi API key and a shared asynchronous workflow.