Muse Wonder Full

Meta's most capable Muse Wonder series model for personal and professional agentic tasks, offered to Muse Code subscribers.

Available for Muse Code subscribers

Compare

Official page Report a correction

6th of 150 on Arena WebDev

The basics

Checked Oct 31

Status: ActiveAvailable for Muse Code subscribers since launch.

Updates

Oct 22

Benchmark

Vals Index: Muse Wonder Full fourth, DeepSeek V4.1 Flash 1017 eighth at $0.66 per test

The Vals Index, updated October 22, 2026, ranks Muse Wonder Full fourth at 66.79% and Muse Wonder sixth at 64.88%. DeepSeek V4.1 Flash 1017 is eighth at 61.26% for $0.66 per test, GLM 5.3 ninth at 59.73%, Grok 4.7 12th at 58.56% and Kimi K3 19th at 52.95%.

Oct 20

Benchmark

Arena: Muse Wonder sixth on the text leaderboard, Wonder Full sixth on WebDev

Arena October 20, 2026 text leaderboard ranks Muse Wonder (Max) sixth at 1502 with 5,099 votes. The October 19 WebDev leaderboard ranks Muse Wonder Full (Max) sixth at 1726 and Muse Wonder (Max) 17th at 1652.

Full changelog, by month

Strengths and weaknesses

Strengths

Active Memory: carrying important context about you across conversations and retrieving relevant details
Active Focus: sustained effort toward challenging goals, including trying alternative approaches and checking in before important actions
Frontier coding for projects from small applications to complex software