Atlas

Compare Models

Muse Wonder vs Qwen4-Max

Qwen4-Max leads by 0.2 capability points (≈95% interval −2.2 to +2.7): statistically tied · P(higher) ≈ 57%

Wins 26 of 51 shared benchmarks (24 lost, 1 tied)

Capability vs. relative cost
Muse WonderQwen4-MaxClosed weightsOpen weights

Benchmark wins24 – 26 · 51 shared, 1 tied

Comparing matching setups only (4 benchmarks excluded) · suites counted once
← Muse WonderQwen4-Max →
Data: Artificial Analysis, Surge AI, Vals AI and 5 othersatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Qwen4-Max minus Muse Wonder.

KradleBenchGames27.611.0−16.6
LMArena Code Arena WebDevSoftware › Frontend & design16481750+13.6
DAYJOB: Finance (October 2026 rescoring)Professional › Finance, accounting & tax38.825.3−13.5
LMArena Agent Arena Net Improvement (Steering Burden)Tool & computer use › General agents0.67.3+13.0
Code Migration v1.1 (Vals Index v2.2 subset)Software › Software engineering69.857.3−12.5
GDP.pdf (November 2026 rescoring)Professional › General knowledge work54.843.6−11.2
Terminal-Bench 4.0 (Artificial Analysis)Software › Software engineering38.448.5+10.1
HANDBOOK.md Agents v1.1Tool & computer use › Customer service & operations29.539.4+9.9
Legal Research BenchProfessional › Legal50.541.3−9.1
ComplexConstraints (November 2026 rescoring)Language & writing › Instruction following67.959.1−8.8
DAYJOB: AccountingProfessional › Finance, accounting & tax30.822.4−8.4
LMArena Agent Arena Praise vs ComplaintTool & computer use › General agents-3.013.0+8.0
EnterpriseBench: CoreCraftTool & computer use › Customer service & operations61.568.7+7.2
DAYJOB: Healthcare (October 2026 rescoring)Health & medicine31.224.4−6.8
LLM Debate BenchmarkLanguage & writing › Language & wordplay16371679+5.9
Humanity's Last Exam DiamondKnowledge & reasoning › General knowledge39.044.9+5.9
CritPt v2 (agentic, avg@5)Science › Physical sciences71.166.0−5.1
Vals Legal IndexProfessional › Legal75.370.7−4.5
AutomationBench-AA 1.0.6Tool & computer use › Customer service & operations68.172.6+4.5
GDP.pdf (Artificial Analysis, All-pass)Professional › General knowledge work30.626.2−4.4
Vibe Code Bench v1.1Software › Frontend & design83.186.7+3.6
LMArena Agent Arena Steering BurdenTool & computer use › General agents1.58.0+3.3
LMArena Agent Arena Confirmed SuccessTool & computer use › General agents2.58.8+3.2
DAYJOB: Law (November 2026 release)Professional › Legal20.617.5−3.1
Artificial Analysis Finance & Accounting IndexProfessional › Finance, accounting & tax51.154.1+3.0
DAYJOB: TaxProfessional › Finance, accounting & tax6.08.8+2.8
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & medicine22.219.4−2.8
Artificial Analysis Legal IndexProfessional › Legal46.949.6+2.7
HLAB-Vals-1.1Professional › Legal23.821.3−2.5
Terminal-Bench-Science 0.1Science › Multidisciplinary25.723.3−2.4
Artificial Analysis Strategy & Ops IndexProfessional › General knowledge work54.556.9+2.4
LiveBench 2.0 (2026-11-06 release)Composite indexes70.773.1+2.4
GDP.xlsxProfessional › General knowledge work21.823.7+1.9
Artificial Analysis Economics IndexComposite indexes49.551.3+1.8
GDPval-AA v2.1Professional › General knowledge work17701809+1.7
Vals Index v2.2Composite indexes64.963.2−1.7
Excel Modeling Benchmark v1.1Professional › Finance, accounting & tax72.570.8−1.7
AA-Briefcase v1.1 EloProfessional › General knowledge work16591686+1.5
FrontierCode 2.0 (Artificial Analysis)Software › Software engineering27.926.5−1.4
AA-Omniscience v1.1 IndexKnowledge & reasoning › General knowledge25.127.7+1.3
Finance Agent v2.1Professional › Finance, accounting & tax65.566.7+1.1
Tax Agent BenchProfessional › Finance, accounting & tax71.270.0−1.1
Terminal-Bench 4.0Software › Software engineering49.548.5−1.0
Vals Finance IndexProfessional › Finance, accounting & tax61.262.0+0.8
Artificial Analysis Intelligence Index v5.0.1Composite indexes43.642.9−0.7
LMArena Agent Arena Tool HallucinationTool & computer use › Tool calling-0.21.2+0.7
Artificial Analysis Engineering IndexScience › Engineering49.650.1+0.5
sudo L7 (October 2026 rescoring)Software › Software engineering16.115.6−0.5
Artificial Analysis Healthcare & Medical IndexHealth & medicine48.748.2−0.5
LMArena Agent Arena Bash ReliabilityTool & computer use › General agents3.43.40.0
DualEntry Accounting AI BenchmarkProfessional › Finance, accounting & tax83.283.20.0
Data: Artificial Analysis, Surge AI, Vals AI and 5 othersatlas.village.dev
Win counts use each benchmark’s declared score direction. Represented suite slices and the derived Capability score are excluded from headline totals. Missing or incompatible setups are excluded, never counted as losses.
Loading Atlas data…