Meta’s Muse Wonder AI models land for Muse Code subscribers

Meta has launched Muse Wonder and Muse Wonder Full for Muse Code subscribers. The models could soon power Muse, with implications for creator AI workflows.

What happened

Meta is expanding its AI lineup with two new models, Muse Wonder and Muse Wonder Full, available for Muse Code subscribers starting today. The company says Wonder can also be accessed through the Meta Model API.

Meta said it eventually plans to power its Muse AI agent with Wonder, starting with subscribers, and will develop specialized models it “can bring to everyone using Muse.” It did not share specifics or provide a timeline for either move.

Meta characterizes the launches as part of an aggressive push to reestablish itself as a major force in AI. The company has poured billions into talent and infrastructure and has been steadily expanding its AI lineup, including with image model Muse Image and a focus on useful AI agents through Muse.

Key facts

• Meta’s new models are Muse Wonder and Muse Wonder Full.
• Both are available for Muse Code subscribers starting today; Wonder is also accessible through the Meta Model API.
• Meta says the models can rival leading models from OpenAI and Anthropic, including GPT-6 Astra, Claude Opus 5.5, and Claude Fable 5.1, based on internal testing on several industry benchmarks.
• The source says it is unclear how the models stack up against Google’s Gemini 4 Argon.
• The launch follows Muse Spark, the first model from Meta’s revamped Superintelligence Labs division, and the Muse AI agent.

Our analysis

This is a subscription-and-API play rather than a broad consumer release. By putting Muse Wonder and Muse Wonder Full behind Muse Code first, Meta is testing adoption with paying users before expanding through its Muse agent and specialized models. That sequencing likely gives Meta a controlled way to collect usage patterns and adjust the models without promising immediate access to everyone.

The competitive framing matters for content tooling. Meta specifically names OpenAI and Anthropic models, but the source notes there are no specifics or timeline for general availability. This suggests creators and operators should treat the announcement as a signal of Meta’s direction, not as an immediate replacement for existing AI workflows. The mention that Google’s Gemini 4 Argon comparison is still unclear also implies the model race remains fluid and benchmark claims should be read as internal testing rather than independent validation.

For social media managers and creators, the most relevant change is the likely convergence of Meta’s AI models with its Muse agent. If Wonder eventually powers Muse, content generation, drafting and agent-based tasks inside Meta’s ecosystem could become more accessible to subscribers first.

What it means for operators

• If you already have Muse Code access, test Muse Wonder and Muse Wonder Full today for drafting, rewriting or ideation before competitors adopt the same workflows.
• If you build or manage AI tooling, evaluate Wonder through the Meta Model API now, because only Wonder is listed as available through the API, not Wonder Full.
• Monitor Meta’s Muse agent updates closely; the company plans to power Muse with Wonder starting with subscribers, so your paid tier may get new capabilities before broader Muse users.
• Do not assume parity with every rival model. Meta names OpenAI and Anthropic models, but the source says it is unclear how the models compare to Google’s Gemini 4 Argon, and there is no independent benchmark data provided.