From the source
Meta released two new frontier models, Muse Wonder and Muse Wonder Full, which score 52 and 56 points respectively on the Artificial Analysis Intelligence Index v4.3, up from 48 for their predecessor Muse Spark 1.3.
The models show significant gains in agentic coding and autonomous problem-solving, with Muse Wonder Full tying Claude Code with Sonnet 5.5 on the Coding Agents Index at 68 points.
However, factual knowledge remains a weak spot, with Muse Wonder Full scoring only 30 points on the AA-Omniscience knowledge test, and API pricing has increased to $4 per million input tokens and $20 per million output tokens for Muse Wonder.