NeuralStackMeta has launched Muse Wonder and Muse Wonder Full, positioning the pair as its most capable artificial intelligence models yet. The release marks a more forceful attempt to compete with OpenAI and Anthropic in an increasingly crowded AI market.
Muse Wonder is expected to power Meta’s Muse personal AI agent, initially for subscribers. The more powerful Muse Wonder Full is available to subscribers of Muse Code, Meta’s AI coding tool.
The launch follows an aggressive year for Meta’s AI push. Chief Executive Mark Zuckerberg has pledged billions of dollars and recruited researchers as the company works to close the gap with leading AI firms.
“Fourteen months ago, we began an overhaul of our AI efforts,” Meta said in a blog post. The company said it had moved from rebuilding its foundations to pushing the frontier of AI development.
Meta’s rollout also builds on the recent launch of its Muse personal AI agent. The company’s successful introduction of that product could create a path to recurring revenue through subscriptions as Meta seeks returns on its substantial AI spending.
Why this matters
The models put Meta’s AI ambitions more directly against OpenAI and Anthropic, rather than limiting the company’s role to infrastructure, research or open-ended experimentation. By connecting Muse Wonder to a personal agent and Muse Wonder Full to a coding product, Meta is tying model development to services that users may pay to access.
That strategy could help Meta distribute its models through products it controls while testing whether advanced AI agents can support a new subscription business. It also raises questions about privacy, particularly as personal agents become more closely integrated into users’ work and daily activities.
Meta’s shares fell 0.7% in premarket trading on Tuesday after the announcement. The company’s next challenge will be turning the launch’s competitive promise into sustained adoption and revenue across Muse and Muse Code.