‹ PublicAI Index
The LLM benchmark aggregator.
Muse Wonder
Meta
Strongest in Cybersecurity (#1 of 46), weakest in Finance (#38 of 161). Above par in 22 of 23 scopes. Among the models it meets almost everywhere, it finishes behind Gemini 4 Argon and Claude Fable 5 and ahead of Claude Opus 4.6 and GPT-5.6 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.
Human preference67.2−1.4#2/3461/1
Human preference67.2−1.4#2/3461/1
Core abilities60.8−4.4#8/2253/4
Data analysis61.4−4.5#4/471/1
General intelligence63.8−5#12/2142/3
Instruction following58.9−14.9#17/621/1
Language56.4−10.2#19/621/1
Coding60.6−5.8#9/1763/6
Agentic coding61.1−5.9#9/1693/5
Agents62−3.9#10/2893/6
Web research61.9−1.8#3/481/2
Knowledge work69.1−2.7#4/1982/2
Long-horizon work57.7−16.3#10/471/1
Computer use55.6−12.3#26/801/2
Reasoning58.4−6.4#18/1893/5
Mathematics60.6−3.4#10/1492/2
Reasoning56.1−10.6#36/1431/3
Professional56.1−7#24/1872/2
Cybersecurity64leads#1/461/1
Legal62.6−3.8#5/1621/1
Biology research44.7−26.4#20/271/1
Document design51.6−15.5#27/471/1
Finance55.8−6.7#38/1611/1
§ 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, left for the other.
§ 3 · Sources
Where the numbers come from
7 publications, 28 figures. Every one links to the page it was read from.
LMArena Text 1509
Artificial Analysis Intelligence Index 52
GDPval-AA 1770
AA-Briefcase 1659
LiveBench 78.1LiveBench · Agentic 53.2LiveBench · Reasoning 90.6LiveBench · Coding 86.2LiveBench · Browser Use 72LiveBench · Data Analysis 81.6LiveBench · Research 88LiveBench · Design 75.1
LiveBench 1.0 81.2LiveBench 1.0 · Reasoning 90.6LiveBench 1.0 · Coding 80.7LiveBench 1.0 · Agentic Coding 62.7LiveBench 1.0 · Mathematics 96.5LiveBench 1.0 · Data Analysis 81.6LiveBench 1.0 · Language 82.9LiveBench 1.0 · Instruction Following 73
Vals · Legal Research Bench 50.48%Vals · Harvey Legal Agent Benchmark 17.08%Vals · Finance Agent 56.64%Vals · BioMysteryBench 67.04%Vals · CyberBench 80.36%Vals · Vibe Code Bench 83.09% (OpenHands)Vals · Code Migration 62.55%Vals · ProofBench 84%
Badge
[](https://publicai.io/model-index/m/muse-wonder)