‹ PublicAI Index
The LLM benchmark aggregator.
Muse Wonder Full
Meta
Strongest in Legal (#2 of 164), weakest in Reasoning (#29 of 190). Above par in 14 of 15 scopes. Among the models it meets almost everywhere, it finishes behind Claude Fable 5 and Gemini 4 Argon and ahead of Muse Wonder and GPT-6 Sol.
§ 1 · Profile
What it is good at
Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.
Core abilities62−3.2#6/2251/4
General intelligence67.3−1.5#4/2141/3
Professional58.8−4.3#7/1881/2
Legal64.1−1.8#2/1641/1
Biology research56.6−11.6#8/281/1
Medical59.5−7#11/1491/1
Cybersecurity57.9−6.1#12/461/1
IT operations48.1−25.7#12/311/1
Finance57.4−5.1#24/1611/1
Agents63−2.9#8/2892/6
Knowledge work70.8−1#3/1982/2
Coding60.3−6.1#10/1761/6
Agentic coding61−6#10/1691/5
Reasoning56.6−8.3#29/1901/5
Mathematics60−4.1#13/1501/2
§ 2 · Head to head
What it beats, and what beats it
The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for Muse Wonder Full, left for the other.
§ 3 · Sources
Where the numbers come from
4 publications, 15 figures. Every one links to the page it was read from.
Artificial Analysis Intelligence Index 56
GDPval-AA 1857
AA-Briefcase 1729
Vals · Legal Research Bench 53.37%Vals · LegalBench 87.42%Vals · Harvey Legal Agent Benchmark 23.33%Vals · Finance Agent 59.06%Vals · MedCode 49.93%Vals · MedScribe 89.33%Vals · BioMysteryBench 73.33%Vals · CyberBench 73.63%Vals · SRE Bench 11.83%Vals · Vibe Code Bench 86% (OpenHands)Vals · Code Migration 61.03%Vals · ProofBench 94%
Badge
[](https://publicai.io/model-index/m/muse-wonder-full)