← Back to Image models
Meta AINotable
Muse Image
Released Jul 7, 2026
The first image generation model from Meta Superintelligence Labs. It powers image creation in Meta AI, Instagram and WhatsApp and pairs with Muse Spark to plan, research and compose before rendering.
- Price per image
- $0.010
- Max resolution
- —
- AA arena Elo
- 1115
- LMArena Elo
- 1274
- Capabilities
- text to image, editing, text rendering, reference images
- Licence
- Proprietary
Strengths & weaknesses
Strengths
- Cheapest model here at a flat $0.01 per image
- Edits and composes from 1-10 input images
- Built-in web/image search lookups at no extra charge
- Plans layout before rendering, including infographics and text in the image
Weaknesses
- Meta says exact text rendering varies run to run
- No mask/inpainting parameter in the API
- Output only up to 1600px
Consumer launch 2026-07-07 (Meta newsroom, updated 07-10). The developer API on the Meta Model API (api.meta.ai/v1) was announced 2026-08-25. The developer blog says output is high-quality "up to 1600px" without giving pixel dimensions, so maxResolution is null. The size parameter sets aspect ratio only. Open weights are not mentioned anywhere, and the model is treated as closed. AA ranks it #8 at 1115 (raw 1114.97), effectively tied with FLUX 3 Image (1114.86).
Links
API model strings
- meta
muse-image-1.0