The Decoder· Matthias Bastian·· 4 d ago
Meta's new frontier model Muse Wonder Full breaks into the AI top tier
Meta's new frontier model Muse Wonder Full breaks into the AI top tier
AI summary
Meta's new Muse Wonder models rank in the top three on six out of ten benchmarks, with the biggest gains in agentic tasks. But the cost per task has nearly tripled.
Selection record
Threshold 76Media and individualsFirst 78Second 82
AdmittedSum of both 160 ≥ twice the threshold 152
- Source tier
- Media and individuals; this tier's threshold is 76
- Pre-filter
- passed：Meta发布新前沿AI模型及评测
- Why it was chosen
- Meta's new frontier model makes big gains on agentic and coding evaluations, but factual knowledge and cost remain weak spots, giving a clear picture of where it really stands.
- Same event
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: The Decoder · the-decoder.com