Meta's Muse Wonder models score 52 and 56 on Artificial Analysis Intelligence Index
Original titleMuse Wonder and Muse Wonder Full: a material jump in intelligence over previous Meta models
Artificial Analysis reports that Meta's Muse Wonder and Muse Wonder Full score 52 and 56 on its Intelligence Index v4.3 at max reasoning effort, up from 48 for Muse Spark 1.3.
The largest gains are in agentic evaluations such as Terminal-Bench 4.0, which rises from 33% to 57% for Muse Wonder Full.
Muse Wonder costs $4.52 per Intelligence Index task, 2.8x its predecessor, while Muse Wonder Full is not priced per token and is available only to Muse Code users.
The article gives benchmark gains alongside cost per task and token use, which helps readers weigh the new models' capability against their price.
Source: Artificial Analysis Articles · artificialanalysis.aiPublished · added here