Atlas

Compare Models

Claude Fable 5.1 vs Muse Spark 1.3

Claude Fable 5.1 leads by 4.6 capability points (≈95% interval +2.3 to +6.8): a clear lead over Muse Spark 1.3 · P(higher) ≈ 100%

Wins 155 of 196 shared benchmarks (37 lost, 4 tied)

Capability vs. relative cost
Claude Fable 5.1Muse Spark 1.3Closed 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 Spark 1.3capability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins155 – 37 · 196 shared, 4 tied

Comparing matching setups only (15 benchmarks excluded) · suites counted once
← Claude Fable 5.1Muse Spark 1.3 →

Claude Fable 5.1vsMuse Spark 1.3wins by categoryData Nov 15, 2026Data: Artificial Analysis, Vals AI, Mercor and 59 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Muse Spark 1.3 minus Claude Fable 5.1.

InferenceBenchCode9.84.5−63.2
BrokenArXiv 08/2026 (Native Harnesses)Math79.820.2−59.5
KernelBench Mega Normalized (RTX PRO 6000)Code64.76.2−58.6
AA-PersonalityBench Transparency indexChat & Writing98.141.0−57.7
BabaIsBench Stage 0: The IntroGames100.046.0−54.0
AA-PersonalityBench Slop indexChat & Writing23.875.3+51.9
ArXivMath 08/2026 (Native Harnesses)Math87.745.6−42.1
ProofBench v1.1Math100.058.0−42.0
BrokenArXiv 05/2026Math78.036.5−41.5
FrontierMath Tier 4 (v2) - PrivateMath87.846.3−41.5
RiemannBenchMath65.628.0−37.6
Document ArenaProfessional Work15151473−37.5
LLM Creative Story WritingChat & Writing3.80.6−36.1
TaxCalcBench TY25Finance & Accounting42.06.0−36.0
CAD ArenaAgents66.231.8−34.3
LMArena Agent Arena Net ImprovementAgents14.13.8−33.8
BrokenArXiv 06/2026Math86.653.2−33.3
Mystery Game PuzzlesGames58.025.0−33.0
ALE-BenchCode28491981−30.8
Terminal-Bench 4.0Code57.927.8−30.1
ChessBench CoherenceGames61.991.2+29.3
Medical Long Context Reasoning (MLCR-AA)Health & Medicine71.143.3−27.8
AA-PersonalityBench Expressiveness indexChat & Writing98.171.4−26.9
FrontierML ($200 wallet)Agents41.916.0−25.9
Humanity's Last Exam DiamondGeneral QA51.325.4−25.9
GameBench 2Games94.068.5−25.5
Code2Video-AA EloCode13251019−24.8
APEX-SWECode63.639.1−24.5
LMArena Agent Arena Net Improvement (Steering Burden)Agents14.82.2−23.9
TextQuests Game ProgressGames56.633.0−23.6
HANDBOOK.mdProfessional Work38.815.4−23.4
AA-PersonalityBench Conviction indexChat & Writing10077.2−23.1
AutoResearchExam (hidden-test AUARC)Agents0.60.4−22.4
HoneyBench v0.2 (hack rate)Safety61.883.6−21.7
RuneBenchGames6.04.6−21.7
KernelBench CUDA Normalized (RTX PRO 6000)Code51.630.3−21.3
LLM Debate BenchmarkChat & Writing17381593−20.6
Terminal-Bench Science 0.1 (Vals)Science34.314.3−20.0
DAYJOB: Data ScienceProfessional Work26.06.0−20.0
AA-AnalystAgent (pass^5)Professional Work57.537.5−20.0
Arena Image-to-WebDevCode17241637−18.8
ChartographyMultimodal46.227.6−18.6
Hedge-BenchFinance & Accounting26.89.1−17.7
sudo L7Code31.113.9−17.2
GBENCH One-shot Coding GScoreGames69.352.7−16.6
Vals RSI Index v1.1 (four tasks)Agents36.119.6−16.5
Vals RSI Index v1.1 (historical five tasks)Agents35.018.7−16.3
NoCoT-Bench NCRI 15.2General QA12898.3−16.3
ITBench-AAAgents49.533.2−16.3
APEX-Agents 1.1Professional Work77.161.3−15.8
FrontierMath Tiers 1-3 (v2) - PrivateMath90.274.4−15.8
AA-3D v1.0 EloCode11391032−15.6
EnterpriseBench: CoreCraftProfessional Work77.461.9−15.5
LMArena Agent Arena Praise vs ComplaintAgents31.61.7−15.0
Gray Swan IPI Attack Success (k=15)Safety1.015.9−14.9
Snorkel Agentic Coding 2.0Code39.625.0−14.6
Last Translation Benchmark (LTBv1-eval, blind)Chat & Writing32.918.4−14.5
ArXivMath 06/2026Math91.076.7−14.2
ArXivMath 05/2026Math87.573.3−14.2
MMMU-Pro Extended (Mercor held-out)Multimodal47.633.8−13.8
DistressBenchSafety71.584.9+13.3
LiveBench 2.0Agents88.074.8−13.2
BuseyBench SVGMultimodal7.05.7−13.2
Westworld Finance Diligence BenchFinance & Accounting58.245.2−13.1
GDPval Extended (Mercor held-out rubric score)Professional Work68.755.9−12.8
MA-BenchProfessional Work49.261.8+12.6
Creative Writing v3 EloChat & Writing21551907−12.6
Snorkel SWE-bench CLICode14.52.2−12.3
Conceptual Reasoning IndexIndexes74.966.2−12.1
Web Clone BenchCode13.31.5−11.7
MysteryMechanismScience47.736.0−11.7
LMCAGeneral QA65.553.9−11.6
AA-PersonalityBench Imagination indexChat & Writing86.798.1+11.5
SpeciEval (Refusal-Aware Retries, Historical)Safety87.298.8+11.5
SpeciEval (Pooled English Runs)Safety87.298.7+11.5
LMArena Code Arena WebDevCode17421655−11.4
Visa Agentic Commerce BenchmarkAgents75.664.4−11.1
Chess Puzzles (task version 2.0)Games48.037.0−11.0
MineBench — Bradley-Terry RatingGames19281747−11.0
ArXivMath 04/2026Math77.566.7−10.8
NoCoT-Bench NCKI (kspine v3)General QA128113−10.8
Immigration BenchLegal75.465.0−10.4
Bug Hunt BenchCode41.030.7−10.3
CursorBench 4.0Code51.841.6−10.2
LegalReasonBenchLegal50.640.4−10.2
WellnessBenchSafety51.061.0+10.0
Professional Reasoning Benchmark - LegalLegal51.661.6+10.0
SnorkelFinance 2.0Finance & Accounting19.610.0−9.6
Excel Modeling BenchmarkFinance & Accounting76.767.4−9.2
DAYJOB: AccountingFinance & Accounting27.218.0−9.2
RSI-Exam 0.1 (Full 88)Agents0.50.4−9.1
CheatBenchSafety45.837.0+8.8
Real-SWECode45.036.3−8.7
Professional Reasoning Benchmark - FinanceFinance & Accounting50.859.5+8.7
DeepFinanceFinance & Accounting45.637.0−8.6
ACCoRDGeneral QA83.775.1−8.6
FrontierCode 1.1 Main (Artificial Analysis)Code49.040.6−8.3
CVDP (517-problem set)Code62.654.3−8.3
Finance Agent v2 - Financial ModelingFinance & Accounting25.333.5+8.2
CadQueryEval (Corrected Geometry)Code76.084.0+8.0
Artificial Analysis Coding Agent Index v1.5Indexes62.254.3−7.9
APEX-AccountingFinance & Accounting61.053.5−7.5
MedXpertQA MM (200-question set)Health & Medicine83.776.3−7.4
CUA-benchAgents13.25.8−7.3
Epoch Capabilities Index (ECI)Indexes167158−7.3
Snorkel Digital Work IndexIndexes23.616.3−7.3
SWE-bench Verified Extended (Mercor held-out)Code61.354.0−7.3
PCT Political Inconsistency (CAIS)Safety33.440.6−7.3
Ramp Accounting BenchFinance & Accounting49.742.5−7.2
Terminal-Bench 2.1Code81.688.8+7.2
FrontierCode 2.0 (Artificial Analysis)Code30.323.1−7.2
SWE-TogetherCode69.362.4−6.9
ComplexConstraintsChat & Writing45.151.9+6.8
Finance Agent v2 - General Qualitative AnalysisFinance & Accounting77.083.4+6.4
DAXBench v2Code91.585.1−6.4
ChessBench EloGames10951201+6.3
AA-LCR Extended (Mercor held-out)Search73.066.9−6.1
LMArena Agent Arena Steering BurdenAgents14.82.6−6.1
DAYJOB: TaxFinance & Accounting9.63.6−6.0
MedXpertQA MM Extended (Mercor held-out)Health & Medicine66.060.1−5.9
Tax Agent BenchFinance & Accounting77.972.4−5.4
HiL-Bench (Human-in-Loop Benchmark)Agents61.566.9+5.4
LMArena Agent Arena Confirmed SuccessAgents14.43.7−5.4
Artificial Analysis Intelligence Index v4.3Indexes53.448.2−5.2
CritPt (avg@5)Science31.126.0−5.1
ChessBench AccuracyGames71.376.0+5.0
DAYJOB: FinanceFinance & Accounting19.814.8−5.0
Extended NYT Connections 940 (Quadratic v1)Games90.085.1−4.9
CritPt v2 (agentic, avg@5)Science72.968.0−4.9
SimpleBenchGeneral QA86.681.8−4.8
Financial Coaching BenchFinance & Accounting75.771.0−4.7
HackerRank-BenchCode47.542.9−4.6
Design Creativity BenchCode48.647.2−4.6
AA-LCR v1.1Search87.583.0−4.5
SnorkelUnderwrite 2.0Finance & Accounting24.019.6−4.4
AA-Briefcase EloAgents16621589−4.2
HardGeoBench Avg DistanceMultimodal8881022−4.1
GDPval-AA v2Professional Work17451662−4.1
LMArena Agent Arena Bash ReliabilityAgents10.12.0−4.0
AA-PersonalityBench Organization indexChat & Writing60.063.8+3.8
AA-PersonalityBench Informativeness indexChat & Writing96.2100+3.8
DAYJOB: LawLegal26.522.8−3.7
GDPval-AA v2.1Professional Work17651681−3.7
LMArena Agent Arena SteerabilityAgents8.41.6−3.4
SciCode 1.0.1 (Artificial Analysis)Code63.159.7−3.4
Humanity's Last ExamGeneral QA46.543.2−3.3
Creative Writing v3 Rubric ScoreChat & Writing16.315.6−3.1
Arena Alignment IndexSafety80.177.1−3.1
Finance Agent v2 - PrecedentsFinance & Accounting32.235.2+3.0
CWE-bench v1 Programmatic Pass@1Cybersecurity58.055.0−3.0
LMArena Agent Arena Bash RecoveryAgents11.66.1−2.8
HardGeoBench <=100miMultimodal52.555.0+2.5
Longform Creative WritingChat & Writing85.382.8−2.5
Harvey LAB-AA v1.1 (hallucination-gated)Legal6.48.9+2.5
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Legal17.415.0−2.4
Realm Financial ReasoningFinance & Accounting47.845.5−2.3
Harvey LAB-AA v1.0 (criterion pass rate)Legal93.395.5+2.2
MentalHealthBenchSafety46.448.6+2.2
Eval Connections One-Shot with TrapsGames82.084.0+2.0
MazeBench ASCIIGames2.00.0−2.0
Shroombench (mushroom identification, top-1)Multimodal51.049.0−2.0
AA-PersonalityBench Warmth indexChat & Writing94.396.2+1.9
Ruby LLM Benchmarks — Program FixerCode76.074.1−1.9
DAYJOB: HealthcareHealth & Medicine9.67.8−1.8
τ³-Banking (Mercor run, bm25_grep retrieval)Agents54.052.2−1.8
Finance Agent v2 - General Quantitative AnalysisFinance & Accounting79.778.0−1.8
Finance Agent v2 - All PassFinance & Accounting47.849.5+1.7
Finance Agent v2 - Earnings AnalysisFinance & Accounting75.577.2+1.7
AutomationBench 1.0.6 (Zapier)Agents22.420.7−1.7
Finance Agent v2 - AdjustmentsFinance & Accounting54.452.8−1.6
Roboflow Vision Evals (September 2026 protocol)Multimodal81.379.8−1.5
SnorkelRevOpsProfessional Work14.012.6−1.4
HardGeoBench Country AccuracyMultimodal61.360.0−1.3
HardGeoBench <=10miMultimodal40.038.8−1.2
CharXiv (descriptive and reasoning combined)Multimodal95.193.9−1.2
Finance Agent v2 - ComparablesFinance & Accounting46.047.2+1.2
LMArena Text Arena EloChat & Writing15031496−1.1
micro1 PharmaBenchHealth & Medicine12.813.9+1.1
DTBenchGeneral QA97.696.5−1.1
OTIS Mock AIME 2024-2025Math100.099.2−0.8
SnorkelManufacturingProfessional Work11.310.5−0.8
APEX-1 (Mercor, mean score)Professional Work73.873.0−0.8
SWE-Marathon v1.1Code45.646.3+0.6
SnorkelLegalLegal35.936.5+0.6
Vision Arena OverallMultimodal12891291+0.6
DeepSWE v1.1Code67.367.8+0.5
GDP.xlsxProfessional Work23.122.6−0.5
Finance Agent v2 - Market AnalysisFinance & Accounting69.569.9+0.4
CharXiv Extended (Mercor held-out)Multimodal82.281.8−0.4
Finance Agent v2 - Disclosure AnalysisFinance & Accounting70.370.1−0.2
GPQA DiamondGeneral QA91.491.3−0.1
LMArena Agent Arena Tool HallucinationAgents0.80.80.0
AIME 2025Math100.0100.00.0
GeneBench-Pro Public Release SubsetScience26.026.00.0
ObviousBenchGeneral QA99.399.30.0
Olam Social Poker EloGames149414940.0

Claude Fable 5.1vsMuse Spark 1.3normalised score marginsData Nov 15, 2026Data: Artificial Analysis, Vals AI, Mercor and 59 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…