Meta Muse Wonder: a frontier model for agents that remembers and keeps pushing
What is it?
Meta Superintelligence Labs released two closed-weight models. The main features are Active Memory, which carries user context across conversations and pulls in what matters, and Active Focus, which keeps working toward a hard goal by trying new approaches while checking in before consequential actions. Standard Muse Wonder is available through the Meta Model API (muse-wonder-1.0) and the Muse Code coding tool. The larger Muse Wonder Full is for Muse Code subscribers only. It takes text, images, audio and video, with a roughly 1M-token context window.
Why does it matter?
At $4 per million input tokens and $20 per million output, it's a new option priced alongside the top tier. On the Artificial Analysis Intelligence Index, Wonder scores 52 and Wonder Full 56Independent, just behind Claude Opus 5.5 at 58Independent. Meta says Wonder Full leads the long-horizon SWE-Marathon benchmark at 57.5Vendor claim. Teams running long agent jobs should put it on their eval shortlist.
Maturity
GAThe standard model is a priced, generally available API, but the strongest variant, Wonder Full, sits behind $50 or $200/month Muse Code plans. By default, Muse Code lets Meta use your inputs to improve its models, so be careful with proprietary code. Zero data retention requires separate approval.
How to track it
- Whether Wonder Full opens up on the API: check the Meta research blog and Meta Model API docs weekly.
- Independent numbers: revisit the Artificial Analysis model page for Wonder Full figures and any rank change.
- Muse Code data policy: watch the official blog for changes to the default use of inputs for training and to zero data retention approval.