Meta releases Muse Wonder, its first frontier AI model
Meta says it put a tremendous amount of effort into making its new Muse Wonder model remember what matters to you, and the company put particular weight on that ability.
Meta releases its first frontier model. Muse Wonder and Muse Wonder Full arrived on Monday.

Artificial Analysis Intelligence Index: Muse Wonder 52 (max), Claude Fable 5.1 53 (max with fallback), GPT-6 Astra 53 (max), GPT-6.1 Sol 52 (max). Meta's own testing puts both models at comparable benchmark scores to Fable 5.1 and GPT-6 Astra.
Two models, one data center. Meta's Superintelligence Lab released Muse Wonder and Muse Wonder Full. Both were trained at the Prometheus data center in central Ohio.
Benchmarks come from Meta's own testing. Meta's model card puts both models at comparable scores to Fable 5.1 and GPT-6 Astra. They are said to do better on complex tasks like coding.
Muse Wonder costs $4 and $20. That is per million input and output tokens through the Meta Model API. Wonder Full is only for Muse Code subscribers, not the API.
Meta puts weight on "active memory". In Meta's testing, Muse Wonder scores higher on MemoryBench than Fable, Opus or Astra. Both models will power the Muse agent.
Meta has released its first frontier model since Scale AI founder Alexandr Wang restructured its AI efforts in August of last year. Meta's Superintelligence Lab put out two models on Monday: Muse Wonder and Muse Wonder Full. Both were trained at Meta's Prometheus data center in central Ohio, which came online earlier this year.
Meta put particular weight on what it calls "active memory," the models' ability to retain and synthesize information a user gives them. In Meta's testing, Muse Wonder scores higher on MemoryBench than Fable, Opus or Astra. Research scientist Kavya Kopparapu said in a launch video that Meta put a tremendous amount of effort into making the model remember what matters to you.
Would you want an AI model that remembers what you tell it?
