Meta releases its first frontier model, Muse Wonder

benchmarksTechCrunch AIOct 12, 2026

Where Muse Spark aimed to be a lightweight model that could perform at low latency, the Wonder models are larger and said to be more capable, with higher benchmark scores and better results on complex tasks like coding.

Impact: HighConfidence: MediumNovelty: Material
benchmarkstechcrunch-aicoding

See the whole story

Read the original on TechCrunch AI

Sens.ai aggregates and assesses published reporting. The assessment above is machine generated; the original source is authoritative.