Meta · 1 credits per image
Muse Image
Meta's first image model, for 1 credit
Muse Image 1.0 is the first image generation model from Meta Superintelligence Labs, released in August 2026. It reasons about complex prompts before drawing and is good at blending several photos into one image.
At 1 credit per image it's one of the cheapest models on AirGallery. Upload two or three photos and ask it to combine them, or write a detailed prompt and let it work out the scene.
Sample images



Every model on AirGallery gets the same three prompts, run once and shown unedited (November 2026). Compare them side by side.
Strengths
- Complex prompts: Reasons about detailed, multi-part descriptions.
- Photo blending: Combines several uploaded photos into one composition.
- Low cost: 1 credit per image.
Best for
- Combining several photos into one scene
- Long, detailed prompts
- Inexpensive experiments
1 credits per image. Credits never expire, no subscription.
Questions
What is Muse Image?
Muse Image 1.0 is Meta Superintelligence Labs' first image generation model, released in August 2026. It uses reasoning to follow complex prompts and can blend multiple photos.
Muse Image or FLUX 3 Klein for cheap drafts?
Both cost 1 credit. Muse handles long prompts and photo blending well; Klein is usually faster. Try both on the same prompt.
Can I use Muse Image without a subscription?
Yes. AirGallery is pay-per-use: buy a credit pack once and each Muse Image generation costs 1 credits. Credits never expire.
Do I need an API key to use Muse Image?
No. AirGallery runs Muse Image for you in the browser. Developers who want it in their own app can call it through the AirGallery API or MCP server with an AirGallery key.
Who owns the images I make with Muse Image?
You do. Images you create are yours to download and use, subject to the model provider's usage policies and AirGallery's terms.