Muse Wonder
Points and comments are a snapshot, not live.
Meta releases Muse Wonder models, claiming top performance on personal and professional agent tasks.
Meta introduces Muse Wonder and Muse Wonder Full, claiming state-of-the-art performance on personal and professional agentic tasks. The series features Active Memory (tracking context across conversations) and Active Focus (sustained effort toward goals). It leads on benchmarks for legal, financial, coding, and safety evaluations, trained at Meta's Prometheus data center. Models are available via Meta Model API and Muse Code. Safety safeguards against misuse are included, per Meta's scaling framework.
What commenters are saying
Commenters are split. Some praise Meta's execution, noting Zuck's bets on Scale and talent hiring have paid off, placing second only to Opus 5.5. Skeptics question benchmark-maxxing, though one notes Meta's own section on measuring it. Others dislike the 'superintelligence' branding, calling it marketing before AGI. Several argue LLMs are becoming a commodity, and Meta's success depends on consumer plans and enterprise adoption.