1/ releasing muse image today — the first image generation model from MSL. it's agentic: pairs with muse spark to reason through your prompt, search the web, and plan before it generates. people get what they meant on the first try. live now in the Meta AI app.
2/ three things i'm excited about under the hood:
- self-refinement (model improves its own output within its chain of thought — emerged during RL, not by design)
- multi-reference composition (many images blended into one coherent generation)
- multi-turn editing (iterate…
3/ it's personal. upload your photos, @-mention friends, pull from what's trending. Easy to share to your story, feed, or group chat. 30+ new ai effects on instagram stories, live in whatsapp direct chats with Meta AI.
it's an exciting start - especially with agentic coding + media generation + personalization, and it's been fun! hopefully we'll see that the joy it brings is itself economically valuable :)
https://meta.ai/share/a/3297cc1c-922b-4d07-ae6f-32a12222134a…