Meta has launched two new models of Artificial Intelligence, Muse Wonder and Muse Wonder Full, which achieve evaluations similar to models of other large companies, such as Anthropic’s Fable 5.1 and OpenAI’s GPT-6 Astra.
© Shutterstock
13/10/2026
by
O meta department dedicated to the development of Artificial Intelligence, the Superintelligence Lab, launched earlier this week those that claim to be its two most advanced models.
These two new models are called Muse Wonder (lighter and faster) and Muse Wonder Full (largest and most capable) and achieve assessments similar to Anthropic's Fable 5.1 and OpenAI's GPT-6 Astra. The idea was to create artificial intelligence models that could handle more complex tasks, such as programming.
“Together, [Muse Wonder and Muse Wonder Full] mark the next step in our climb a great leap towards personal superintelligence”, can be read in the Meta statement about the launch of these models.
Remember that these are the most advanced models launched by the Artificial Intelligence Division of the Meta of this one that it became led by Alexandr Wang, best known for being founder of Scale AI.
It is known that these new models of Artificial Intelligence have been specially developed to retain information and allow users to perform tasks over time.
“ We put a tremendous amount of effort into ensuring that your model can remember what is important to you over days, weeks, months and beyond.”, said one of the leaders for the development of these new Meta Artificial Intelligence models, Kavya Kopparapu, according to the techCrunch site.