Meta Muse Image generates high-fidelity images with strong instruction following, detailed typography, plots, and QR codes. Upload up to 10 reference images to switch automatically to edit mode.
Added Aug 27, 2026
Approx. Price
$0.010 per image
Model Type
both
Settings
Generation controls available for this model.
Images Per Run
Up to 10
Output images
Input Images
Up to 10
Reference/edit images accepted • Route max 30 MB
Output Sizes
Aspect Ratio
Default
auto
Number of Images
Default
1
Output Format
Default
webp
Options (3)
WebP, PNG, JPEG
Choose the generated image format.
Resolution
Default
auto
Options (10)
auto (Auto (Muse chooses)), 21:9 (21:9), 16:9 (16:9), 4:3 (4:3) +6 more
Benchmarks
Benchmarks
Human preference benchmarks sourced from LMArena.
Text to Image
#3 / 75
Arena Score
1283.4
Votes
14,510
Confidence Interval
1276.6 - 1290.2
Image Edit
#2 / 52
Arena Score
1404.7
Votes
47,573
Confidence Interval
1399.0 - 1410.3
Published 2026-08-07 · Matched as muse-image
LMArena DatasetExamples
Loading examples…
Related image models
Compare Muse Image with similar models from the same provider or model family.
Boogu Image
fal-ai/boogu-imageBoogu Image generates stylized images from text prompts through FAL.
Boogu Image Edit
fal-ai/boogu-image/editBoogu Image Edit transforms one uploaded image from natural language instructions.
Bernini R Edit Image
fal-ai/bernini-r/edit-imageBernini R edits an uploaded image from natural language instructions.
Luma UNI-1
luma/agent/uni-1/v1Luma's multimodal image model for high-fidelity text-to-image generation and prompt-guided image edits in one model. Supports reference images, optional web grounding, manga style, and PNG/JPEG outputs.
Luma UNI-1 Max
luma/agent/uni-1/v1/maxMaximum-fidelity UNI-1 generation and editing for richer detail, stronger prompt adherence, reference following, and high-quality final stills.
ImagineArt 2.0 Edit Preview
imagineart/imagineart-2.0-edit-preview/image-to-imagePrompt-guided image editing at 2K resolution, preserving fine detail and realism while applying targeted changes across one or more reference images.