Meta Muse Image is the first image generation model from Meta Superintelligence Labs. Unlike models that map a prompt straight to pixels, Muse Image reasons through the request first — planning composition, then generating — and reviews its own draft inside its chain of thought, making local edits or full regenerations when something's off. The practical result is more reliable text-in-image, layouts, and multi-step prompts than a pure diffusion model tends to produce.
| Developer | Meta (Meta Superintelligence Labs) |
|---|---|
| Model ID | meta/muse-image |
| Release Date | August 26, 2026 |
| Modalities | Text (input) → Image (output) |
| Pricing | $0.013 per image |
Plans composition before rendering instead of mapping the prompt straight to pixels, which holds together better on requests involving spatial relationships.
Reviews its own draft inside its chain of thought — making a local edit if a small detail is off, or a full regeneration if larger parts are wrong.
Reads and renders text inside images accurately — useful for infographics, UI mockups, and product labels.
Anchor a series of generations on a small set of reference images to keep character, style, and setting consistent across outputs.
Invokes tools during generation, refining its own output as it works rather than producing a single unreviewed pass.
Ranks in the top tier of human-preference leaderboards for text-to-image generation at launch.
Generate on-brand imagery, ads, and product mockups from text prompts, with legible in-image text and consistent styling across a series.
Accurate text rendering makes it a practical fit for step-by-step graphics and interface mockups — areas pure diffusion models tend to fumble.
Anchored composition keeps character, style, and setting consistent from one generation to the next, useful for storyboards or campaign sets.
Measured on the AI/ML API gateway across live traffic. Figures are medians unless stated. Last updated 2026-10-12.
Muse Image is documented for general image generation and editing, including creating images from text prompts, editing existing images while keeping the rest intact, and refining results through an ongoing conversation. It is also described as suited to instruction-following creative work, such as following complex prompts faithfully and composing images from multiple references.
It is documented for grounded, real-world image creation, including redesigning a room with real products from the web or Marketplace. It is also described as useful for advertiser workflows, including producing on-brand ad variations, swapping styles, and creating campaign creative with fewer iterations.
It is documented for social and app-integrated image workflows, including generating images in direct chats, powering Instagram Stories effects, and creating edits that use photos as inputs. Meta also describes it as capable of using web search to ground generated images in factual and real-time information and visual references.
import requests r = requests.post( "https://api.aimlapi.com/v1/images/generations", headers={"Authorization": "Bearer " + AIMLAPI_KEY}, json={ "model": "meta/muse-image", "prompt": "A cat holding a sign that says hello" }, ) print(r.json())
const r = await fetch("https://api.aimlapi.com/v1/images/generations", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"model": "meta/muse-image",
"prompt": "A cat holding a sign that says hello"
}),
});
console.log(await r.json());
curl -X POST https://api.aimlapi.com/v1/images/generations \ -H "Authorization: Bearer $AIMLAPI_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"meta/muse-image","prompt":"A cat holding a sign that says hello"}'
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Output | |
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
Meta Muse Image This page | Image generation from text prompts | |||
| Image generation at scale | ||||
| Image generation at scale | ||||
| Image generation at scale | ||||
| Image generation at scale |
Meta Muse Image takes text as input and returns image.
Meta Muse Image became available on August 26, 2026.
Meta Muse Image is priced at $0.013 / gen.
Meta Muse Image is billed per generation — a fixed charge per output rather than by prompt length.
Meta Muse Image was built by Meta.
Send a request to https://api.aimlapi.com/v1/images/generations with meta/muse-image as the model id.
Yes. Meta Muse Image is served through AI/ML API, so the same key and endpoint format used for other models applies.
Muse Image renders fine details like text, plots and QR codes accurately, according to its description.
Meta describes Muse Image as having faithful instruction-following, meaning it closely follows the details given in a prompt.
