Muse Image Generator & Editor

Generate or edit photos with Meta Muse Image. Keep type readable on posters and packaging, or restyle a scene with up to 10 photo references. Muse sets the pixel size from the ratio.

About 2.5 megapixels
Up to 10 photo refs
Text and edit
Meta · Muse Image docs ↗

Example prompts for Muse Image

Enamel pins

Prompt

Enamel pins on linen over wood. A scalloped portrait pin of a smiling person with dark wavy hair sits in the center. Readable pins say JOURNAL, DAILY DOSE, OUT & RESET, SLOW · BREATHE, and @JUNE.MOON beside an Instagram mark. Also a leaf, a pink heart, a camera, a backpack, a mountain shield, and a coffee cup. 1:1.

Key Features of Muse Image

Image generation, precise editing, anchored composition, and precise typography. Prefill any row; nothing runs until you generate.

Image generation

Write the picture in the prompt. Muse Image generates it on text-to-image.

Diver in a black wetsuit swimming over a coral reef, with sun shafts, rising bubbles, and small fish

How to Use Muse Image

1
1. Start from text or photos
Open Muse Image with a text prompt, or attach one to ten owned photos for image edit. Prefill from a Key Feature or Example if you want. Nothing runs until you generate. Generate stays disabled on edit until a photo is attached.
Open Muse Image with a text prompt, or attach one to ten owned photos for image edit. Prefill from a Key Feature or Example if you want. Nothing runs until you generate. Generate stays disabled on edit until a photo is attached.
2
2. Write the scene, or the one change
For a new photo, name the subject, the type that must be readable, and the light. For an edit, name the change and what must stay. Refer to uploads as image 1 and image 2 when you need a role.
For a new photo, name the subject, the type that must be readable, and the light. For an edit, name the change and what must stay. Refer to uploads as image 1 and image 2 when you need a role.
3
3. Choose a ratio, then generate
Pick Auto or a preset ratio, including 9:21. Muse sets the pixel size. Confirm the credit estimate and generate. Review the photo, then download or continue on image edit. There is no resolution or quality menu.
Pick Auto or a preset ratio, including 9:21. Muse sets the pixel size. Confirm the credit estimate and generate. Review the photo, then download or continue on image edit. There is no resolution or quality menu.

Create with Muse Image ↑

Muse Image Pricing

Estimates below use the 1-credit image. Generate and edit share that MixVio rate. Confirm the exact reservation before you generate.

Compare Muse ImageHow Muse Image compares to other models

Pick by the job: readable type with up to 10 photo references, Flare or Sunburst keep-and-change, native 2K and 4K, or dense multilingual layouts. Specs below are the live MixVio routes.

Muse Image

This model

Readable type, up to 10 photo refs, and about 2.5 megapixels

Provider-safe PNG1 credit per output

GPT Image 2.5 →OpenAI

Keep-and-change edits, reference fidelity, and Flare or Sunburst

1K–4K PNG1–8 credits per output

FLUX 3 Image →Black Forest Labs

Native 2K and 4K photos, readable text, and up to 10 photo refs

1K–4K PNG3–43 credits per output

Attribute	This modelMuse Image	OpenAIGPT Image 2.5 →	Black Forest LabsFLUX 3 Image →	AlibabaQwen Image 3 →
Best for	Readable type, up to 10 photo refs, and about 2.5 megapixels	Keep-and-change edits, reference fidelity, and Flare or Sunburst	Native 2K and 4K photos, readable text, and up to 10 photo refs	Dense layouts, readable multilingual text, and up to three photo refs
Text rendering	Strong	Strong	Strong	Excellent
Image editing	Strong	Excellent	Strong	Strong
Multi-image references	Excellent	Strong	Excellent	Medium
Photorealism	Strong	Excellent	Excellent	Strong
Speed	Medium	Excellent	Medium	Medium
Max resolution	—	1K–4K	1K–4K	1K–2K
Credits	1 credit per output	1–8 credits per output	3–43 credits per output	3–6 credits per output

Stay on Muse Image for readable type, up to 10 photo references, and a pixel size Muse picks, about 2.5 megapixels. Stay on GPT Image 2.5 for Flare or Sunburst, 1K–4K, and up to 16 references. Stay on FLUX 3 Image for native 2K and 4K. Stay on Qwen Image 3 for dense multilingual layouts at 1K or 2K with up to three photos. Reasoning toggles, search, code, Content Seal, custom ratios, and output batches are not MixVio controls on this family. Speed is the MixVio catalog tier, not a MixVio latency benchmark.

Muse ImageGPT Image 2.5FLUX 3 ImageQwen Image 3

Muse Image FAQs

What is Muse Image on MixVio?

Muse Image is Meta’s image model for generating and editing photos. Text-to-image takes a prompt. Image edit takes a prompt plus 1 to 10 owned photos. This page is separate from Muse Video. Muse Image ↗

Which aspect ratios are supported?

Auto, 21:9, 16:9, 4:3, 3:2, 1:1, 2:3, 3:4, 9:16, and 9:21. Auto is the default and lets Muse choose the frame. 9:21 is a preset, not a free-form width and height picker.

Does Muse Image take a resolution?

No. There is no 1K, 2K, or 4K control. Muse picks the pixel size from the ratio. The model card describes that size as about 2.5 megapixels. Returned pixels can vary.

How many photo references can I attach?

Text-to-image takes no photos. Image edit takes 1 to 10 JPEG, PNG, or WebP photos. Order is image 1 through image 10. MixVio does not publish a pixel-size gate for these photos. Adding more photos does not change the credit quote.

How many credits does Muse Image use?

Muse Image uses 1 credit per photo. Generate and edit share that rate, and the reference count does not change it. The generator shows the reservation before you submit. Failed runs cost no credits.

Is Muse Image free?

It is available on the Free plan when your balance covers the quote. Generation is not zero-credit. Free delivery limits such as a watermark may still apply. Failed runs cost no credits.

How does Muse Image compare with GPT Image 2.5, FLUX 3 Image, and Qwen Image 3?

Stay on Muse Image for readable type, up to 10 photo references, and a pixel size Muse picks. Stay on GPT Image 2.5 for Flare or Sunburst, 1K–4K, and up to 16 references. Stay on FLUX 3 Image for native 2K and 4K. Stay on Qwen Image 3 for dense multilingual layouts at 1K or 2K with up to three photos. Specs in the comparison table are the live MixVio routes.

Does MixVio expose reasoning, search, or a custom ratio?

No. Meta documents agentic behavior, including reasoning, search, and code, as model behavior. MixVio does not expose those as toggles, and it does not expose Content Seal, a free-form ratio, or a batch of 2 to 10 images. One JPEG comes back per run.

Can I use Muse Image photos commercially?

Paid-plan outputs may be used commercially, subject to the Terms of Service and current Meta terms. You must have rights to every prompt, reference, face, and brand. Free-plan outputs are for evaluation. MixVio does not guarantee exclusivity or trademark clearance.

Are my prompts and outputs private on MixVio?

Yes. Prompts are routed server-side, and final media is archived privately to your workspace. Keep private prompts and asset IDs out of shared links.

Create and edit photos with Muse Image.
Private. Credits shown first.

Start from a prompt or up to 10 owned photos, choose a ratio, and let Muse set the size. Then generate with Muse Image selected.