Atlas

Compare Models

Claude Fable 5.1 vs Muse Wonder

Claude Fable 5.1 leads by 2.2 capability points (≈95% interval −0.2 to +4.6): statistically tied · P(higher) ≈ 96%

Wins 115 of 170 shared benchmarks (54 lost, 1 tied)

Capability vs. relative cost
Claude Fable 5.1Muse WonderClosed 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 Wondercapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins115 – 54 · 170 shared, 1 tied

Comparing matching setups only (18 benchmarks excluded) · suites counted once
← Claude Fable 5.1Muse Wonder →

Claude Fable 5.1vsMuse Wonderwins by categoryData Nov 15, 2026Data: Artificial Analysis, Vals AI, Mercor and 47 othersVillage LLM Atlasatlas.village.dev

Margins

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

BioSecBench-Refusal V1 Clean RerunsSafety5.655.3+49.7
DeepResearch Bench II (Mercor 130-task set)Search55.010.8−44.2
AA-PersonalityBench Transparency indexChat & Writing98.156.2−42.3
LisanBench Path LengthGames1510027973+40.0
LisanBench Difficulty-WeightedGames45738303+39.9
VCT-Refusal (CAIS)Safety98.960.2−38.7
ChessBench CoherenceGames61.9100.0+38.1
CheatBenchSafety45.812.7+33.1
AA-PersonalityBench Organization indexChat & Writing60.090.5+30.8
AA-PersonalityBench Slop indexChat & Writing23.854.3+30.8
Code2Video-AA EloCode1325997−26.6
EBR-bench v4 (card ban, single agent)Games57.131.4−25.7
LMArena Agent Arena Net Improvement (Steering Burden)Agents14.81.8−24.6
Build-A-Chip-Bench (20-task index run)Agents23.50.3−23.2
FrontierML ($200 wallet)Agents41.919.3−22.6
TextQuests Average Cumulative HarmSafety27.120.8+21.6
AA-PersonalityBench Conviction indexChat & Writing10079.1−21.2
LLM Creative Story WritingChat & Writing3.82.0−20.3
MysteryMechanismScience47.727.5−20.3
BullshitBench v2 LeaderboardSafety77.057.0−20.0
TaxCalcBench TY25Finance & Accounting42.022.0−20.0
HoneyBench v0.2 (hack rate)Safety61.881.0−19.2
HiL-Bench (Human-in-Loop Benchmark)Agents61.579.7+18.2
HANDBOOK.md Agents v1.1Professional Work47.129.5−17.6
HardGeoBench <=10miMultimodal45.027.5−17.5
FrontierMath Tier 4 (v2) - PrivateMath87.870.7−17.1
LMArena Agent Arena Praise vs ComplaintAgents31.6-1.8−16.7
NoCoT-Bench NCRI 15.2General QA12897.7−16.6
SnorkelManufacturingProfessional Work11.327.8+16.5
OpenWorldEngineer (Mean@8)Code29.112.7−16.4
GameBench 2Games94.077.8−16.2
HardGeoBench Avg DistanceMultimodal8451369−16.1
ProofBench v1.1Math100.084.0−16.0
EnterpriseBench: CoreCraftProfessional Work77.461.5−15.9
Furniture Assembly BenchmarkMultimodal70.054.2−15.8
RuneBenchGames6.05.0−15.6
sudo L7Code31.115.6−15.5
HardGeoBench <=100miMultimodal53.838.8−15.0
DistressBenchSafety71.585.9+14.3
Humanity's Last Exam DiamondGeneral QA51.337.3−14.0
NoCoT-Bench NCKI (kspine v3)General QA128109−13.1
LMArena Code Arena WebDevCode17421649−12.2
RiemannBench (November 2026 rescoring)Math71.259.2−12.0
GBENCH One-shot Coding GScoreGames69.358.1−11.2
Bug Hunt BenchCode41.030.2−10.8
SpeciEval (Pooled English Runs)Safety87.297.5+10.3
DAYJOB: LawLegal26.536.6+10.1
Conceptual Reasoning IndexIndexes74.967.7−9.9
LiveBench 2.0Agents88.078.1−9.9
AA-PersonalityBench Expressiveness indexChat & Writing98.188.6−9.6
Web Clone BenchCode13.33.7−9.6
Excel Modeling BenchmarkFinance & Accounting76.767.1−9.6
DAYJOB: Data ScienceProfessional Work26.016.7−9.3
SWE Atlas v1.1 - OverallCode74.164.8−9.3
APEX-SWECode63.654.6−9.0
Eval Connections One-Shot with TrapsGames82.073.0−9.0
TextQuests Game ProgressGames56.647.6−9.0
OSWorld 2.1 full - reported binary completion, step limit unspecifiedAgents43.134.3−8.8
FrontierCode 2.0 - Pass RateCode37.628.8−8.8
FrontierCode 2.0 - ScoreCode34.025.3−8.7
LMCAGeneral QA65.556.8−8.6
DAYJOB: HealthcareHealth & Medicine9.618.0+8.4
SciCode (main problems)Science24.215.8−8.4
EnigmaEvalGeneral QA44.236.1−8.1
OSWorld 2.1 full - reported partial reward, step limit unspecifiedAgents80.772.6−8.1
FrontierMath Tiers 1-3 (v2) - PrivateMath90.282.1−8.1
Professional Reasoning Benchmark - FinanceFinance & Accounting50.858.7+7.9
ACCoRDGeneral QA83.776.2−7.5
Creative Writing v3 EloChat & Writing21552007−7.5
ComplexConstraints (November 2026 rescoring)Chat & Writing60.467.9+7.5
SWE-Marathon - release-reported version unspecifiedCode45.653.1+7.5
DAXBench v2Code91.584.0−7.4
Finance Agent v2 - Market AnalysisFinance & Accounting69.562.1−7.4
RWE-benchAgents7.00.0−7.0
AutomationBench-AA - release-reported strict pass rateAgents32.139.1+7.0
Mystery Game PuzzlesGames58.051.0−7.0
SWE-Marathon v1.1Code45.652.5+6.9
CritPt v2 (agentic, avg@5)Science77.770.9−6.9
Professional Reasoning Benchmark - LegalLegal51.658.4+6.8
Tax Agent BenchFinance & Accounting77.971.2−6.7
Visa Agentic Commerce BenchmarkAgents75.682.2+6.7
MindCube TinyMultimodal84.591.1+6.6
HardGeoBench Country AccuracyMultimodal58.852.5−6.2
Finance Agent v2 - General Qualitative AnalysisFinance & Accounting77.070.8−6.2
LMArena Agent Arena Steering BurdenAgents14.82.6−6.1
CUA-benchAgents13.27.2−6.0
Immigration BenchLegal75.469.6−5.8
AA-PersonalityBench Warmth indexChat & Writing94.3100+5.8
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Legal17.423.0+5.6
SpatialViz-BenchMultimodal77.983.5+5.6
Toloka Arena (v3)Agents71.065.7−5.3
ERQAMultimodal70.375.3+5.0
SnorkelFinance 2.0Finance & Accounting19.615.0−4.6
LMArena Agent Arena Confirmed SuccessAgents14.45.4−4.5
micro1 PharmaBenchHealth & Medicine12.817.2+4.4
MMMU-Pro Extended (Mercor held-out)Multimodal47.643.2−4.4
BuseyBench SVGMultimodal7.06.6−4.3
Excel Modeling Benchmark v1.1Finance & Accounting76.772.5−4.2
Epoch Capabilities Index (ECI)Indexes167162−4.1
WellnessBenchSafety51.055.0+4.0
Snorkel Agentic Coding 2.0Code39.643.6+4.0
SnorkelRevOpsProfessional Work14.018.0+4.0
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine18.322.2+3.9
AA-PersonalityBench Imagination indexChat & Writing86.790.5+3.8
MineBench — Bradley-Terry RatingGames19281868−3.6
Harvey LAB-AA v1.1 (hallucination-gated)Legal6.410.0+3.6
DAYJOB: TaxFinance & Accounting9.66.0−3.6
DAYJOB: AccountingFinance & Accounting27.230.8+3.6
Machiavelli Deception (CAIS)Safety82.483.6−3.5
MMLU ProGeneral QA93.089.6−3.4
Real-SWE 1.1Code45.642.4−3.2
GDP.pdfProfessional Work27.630.8+3.2
Snorkel SWE-bench CLICode14.511.4−3.1
Finance Agent v2 - Disclosure AnalysisFinance & Accounting70.367.3−3.0
HPCT Refusal (CAIS)Safety100.097.0−3.0
ChessBench EloGames10951046−2.9
CritPt (avg@5)Science31.128.3−2.9
Finance Agent v2 - Financial ModelingFinance & Accounting25.322.5−2.8
Snorkel Digital Work IndexIndexes23.626.4+2.8
SciCode 1.0.1 (Artificial Analysis)Code63.160.3−2.8
LMArena Agent Arena Bash ReliabilityAgents10.14.8−2.6
CharXiv Extended (Mercor held-out)Multimodal82.279.6−2.6
JobBench - release-reported mean task rewardProfessional Work73.170.5−2.6
FrontierCode 1.1 Main (Artificial Analysis)Code49.046.5−2.5
IntPhys 2Multimodal67.770.2+2.5
SnorkelLegalLegal35.938.3+2.4
DTBenchGeneral QA97.695.2−2.4
HackerRank-BenchCode47.549.7+2.2
Finance Agent v2 - General Quantitative AnalysisFinance & Accounting79.777.6−2.1
τ³-Banking (Mercor run, bm25_grep retrieval)Agents54.051.9−2.1
MazeBench ASCIIGames2.00.0−2.0
Finance Agent v2 - ComparablesFinance & Accounting46.047.9+2.0
Financial Coaching BenchFinance & Accounting75.773.8−1.9
AA-PersonalityBench Informativeness indexChat & Writing96.294.3−1.9
DualEntry Accounting AI BenchmarkFinance & Accounting85.183.2−1.9
Finance Agent v2 - All PassFinance & Accounting47.845.9−1.8
Finance Agent v2 - AdjustmentsFinance & Accounting54.452.7−1.7
GDPxProfessional Work8.56.8−1.7
Snorkel WorkplaceAgentsProfessional Work15.817.4+1.6
DAYJOB: FinanceFinance & Accounting19.821.3+1.5
Artificial Analysis Coding Agent Index v1.5Indexes62.263.7+1.4
LMArena Text Arena EloChat & Writing15031511+1.4
AA-LCR Extended (Mercor held-out)Search73.074.4+1.4
CursorBench 4.0Code51.850.4−1.4
DeepFinanceFinance & Accounting45.644.2−1.4
APEX-1 (Mercor, mean score)Professional Work73.875.1+1.3
GDPval Extended (Mercor held-out rubric score)Professional Work68.770.0+1.3
Finance Agent v2 - Earnings AnalysisFinance & Accounting75.576.8+1.3
MissionReportProfessional Work63.362.1−1.2
MMMU ProMultimodal86.184.9−1.2
ChessBench AccuracyGames71.370.2−1.2
Humanity's Last ExamGeneral QA46.547.6+1.1
MASKSafety85.984.9−1.0
SnorkelUnderwrite 2.0Finance & Accounting24.023.0−1.0
Ramp Accounting BenchFinance & Accounting49.748.7−1.0
BioTIER RefusalSafety99.598.7−0.8
LMArena Agent Arena Tool HallucinationAgents0.8-0.5−0.7
Prophet Arena Arena Score (Agentic, OpenCode)Forecasting86.085.4−0.7
AA-Briefcase v1.1 EloAgents16711659−0.7
CharXiv (descriptive and reasoning combined)Multimodal95.194.5−0.6
OTIS Mock AIME 2024-2025Math100.099.4−0.6
GDPval (Mercor 213-task rubric score)Professional Work86.687.0+0.4
APEX-Accounting 1.1Finance & Accounting66.767.1+0.4
GDPval-AA v2.1Professional Work17651770+0.3
Creative Writing v3 Rubric ScoreChat & Writing16.316.2−0.2
Prophet Arena Arena Score (Fixed Context)Forecasting85.085.2+0.1
Finance Agent v2 - PrecedentsFinance & Accounting32.232.1−0.1
APEX-Agents 1.1Professional Work77.177.0−0.1
PCT Political Inconsistency (CAIS)Safety33.433.3+0.1
Dual-Arm Sim (task completion)Agents0.00.00.0

Claude Fable 5.1vsMuse Wondernormalised score marginsData Nov 15, 2026Data: Artificial Analysis, Vals AI, Mercor and 47 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…