Atlas

Compare Models

Claude Fable 5.1 vs Muse Wonder Full

Claude Fable 5.1 leads by 1.1 capability points (≈95% interval −1.4 to +3.6): statistically tied · P(higher) ≈ 80%

Wins 29 of 54 shared benchmarks (25 lost)

Capability vs. relative cost
Claude Fable 5.1Muse Wonder FullClosed weightsOpen weights

Each line runs through one model’s priced reasoning settings, cheapest to most expensive; grey points are other models’ priced settings in the same range. Results whose reasoning effort the source didn’t report are left out wherever a model has named effort levels. Capability and relative cost are the published estimates (relative cost is cost per task as a multiple of GPT-5 (high)); hover a setting, or use the arrow keys, to set it beside the other model’s same setting.

Claude Fable 5.1vsMuse Wonder Fullcapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins29 – 25 · 54 shared

Comparing matching setups only (12 benchmarks excluded) · suites counted once
← Claude Fable 5.1Muse Wonder Full →

Claude Fable 5.1vsMuse Wonder Fullwins by categoryData Nov 15, 2026Data: Vals AI, Artificial Analysis, Arena and 9 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Muse Wonder Full minus Claude Fable 5.1.

LisanBench Path LengthGames1510032190+53.1
LisanBench Difficulty-WeightedGames45739337+51.0
AutomationBench-AA - release-reported strict pass rateAgents32.145.4+13.3
MysteryMechanismScience47.735.1−12.6
SWE-Marathon - release-reported version unspecifiedCode45.657.5+11.9
Vals RSI Index v1.1 (four tasks)Agents36.124.4−11.7
LMArena Agent Arena Net Improvement (Steering Burden)Agents14.89.0−11.1
LMArena Agent Arena Praise vs ComplaintAgents31.615.4−8.1
RiemannBench (November 2026 rescoring)Math71.263.2−8.0
SWE-Marathon v1.1Code45.653.1+7.5
FrontierCode 2.0 (Artificial Analysis)Code30.337.8+7.5
Harvey LAB-AA v1.1 (hallucination-gated)Legal6.413.1+6.7
ProofBench v1.1Math100.094.0−6.0
Artificial Analysis Coding Agent Index v1.5Indexes62.268.0+5.8
GDP.pdfProfessional Work27.633.2+5.6
Excel Modeling BenchmarkFinance & Accounting76.771.9−4.8
Finance Agent v2 - Market AnalysisFinance & Accounting69.564.9−4.6
Robo-benchAgents5.810.4+4.6
Immigration BenchLegal75.471.2−4.2
FrontierCode 1.1 Main (Artificial Analysis)Code49.044.9−4.1
GDPval-AA v2.1Professional Work17651857+4.1
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine18.322.2+3.9
AA-LCR v1.1Search87.583.7−3.8
OSWorld 2.1 full - reported binary completion, step limit unspecifiedAgents43.139.4−3.7
MedCodeHealth & Medicine53.549.9−3.6
AA-Briefcase v1.1 EloAgents16711729+3.2
Finance Agent v2 - Earnings AnalysisFinance & Accounting75.578.6+3.1
Tax Agent BenchFinance & Accounting77.975.0−2.9
Finance Agent v2 - PrecedentsFinance & Accounting32.235.0+2.8
Finance Agent v2 - General Qualitative AnalysisFinance & Accounting77.074.2−2.8
LMArena Agent Arena Steering BurdenAgents14.89.6−2.6
LMArena Code Arena WebDevCode17421723−2.4
OSWorld 2.1 full - reported partial reward, step limit unspecifiedAgents80.778.4−2.3
CritPt v2 (agentic, avg@5)Science72.975.1+2.3
Finance Agent v2 - AdjustmentsFinance & Accounting54.456.7+2.3
JobBench - release-reported mean task rewardProfessional Work73.175.3+2.2
MedScribeHealth & Medicine91.389.3−2.0
Excel Modeling Benchmark v1.1Finance & Accounting76.775.0−1.7
CritPt (avg@5)Science31.132.9+1.7
LMArena Agent Arena Confirmed SuccessAgents14.411.1−1.6
Finance Agent v2 - Financial ModelingFinance & Accounting25.323.7−1.6
MASKSafety85.987.3+1.4
Finance Agent v2 - Disclosure AnalysisFinance & Accounting70.371.6+1.3
Finance Agent v2 - All PassFinance & Accounting47.849.0+1.3
LegalBenchLegal88.587.4−1.1
Humanity's Last Exam (Text-Only)General QA54.653.6−1.0
Bug Hunt BenchCode41.040.0−1.0
LMArena Agent Arena Bash ReliabilityAgents10.18.3−0.9
Finance Agent v2 - ComparablesFinance & Accounting46.046.8+0.8
Financial Coaching BenchFinance & Accounting75.775.0−0.7
GDPxProfessional Work8.59.0+0.5
SciCode 1.0.1 (Artificial Analysis)Code63.162.6−0.5
LMArena Agent Arena Tool HallucinationAgents0.80.0−0.4
Finance Agent v2 - General Quantitative AnalysisFinance & Accounting79.780.1+0.3

Claude Fable 5.1vsMuse Wonder Fullnormalised score marginsData Nov 15, 2026Data: Vals AI, Artificial Analysis, Arena and 9 othersVillage LLM Atlasatlas.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…