Atlas

Compare Models

Gemini 3.7 Flash vs Muse Spark 1.3

Muse Spark 1.3 leads by 1.8 capability points (≈95% interval −0.2 to +3.8): statistically tied · P(higher) ≈ 96%

Wins 112 of 176 shared benchmarks (54 lost, 10 tied)

Capability vs. relative cost
Gemini 3.7 FlashMuse 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.

Gemini 3.7 FlashvsMuse Spark 1.3capability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins54 – 112 · 176 shared, 10 tied

Comparing matching setups only (6 benchmarks excluded) · suites counted once
← Gemini 3.7 FlashMuse Spark 1.3 →

Gemini 3.7 FlashvsMuse Spark 1.3wins by categoryData Nov 15, 2026Data: Vals AI, Arena, Artificial Analysis and 37 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 Gemini 3.7 Flash.

Harvey's Legal Agent Benchmark - Data Privacy CybersecurityLegal0.060.0+60.0
BabaIsBench Stage 0: The IntroGames100.046.0−54.0
BrokenArXiv 06/2026Math11.153.2+42.1
Harvey's Legal Agent Benchmark - Bankruptcy RestructuringLegal0.040.0+40.0
Harvey's Legal Agent Benchmark - Corporate GovernanceLegal20.060.0+40.0
Harvey's Legal Agent Benchmark - Emerging Companies Venture CapitalLegal0.040.0+40.0
Harvey's Legal Agent Benchmark - ImmigrationLegal0.040.0+40.0
BrokenArXiv 04/2026Math19.349.2+29.9
Chess Puzzles (task version 2.0)Games66.037.0−29.0
CyberBench v1.1 OverallCybersecurity43.872.7+29.0
Medical Long Context Reasoning (MLCR-AA)Health & Medicine15.043.3+28.3
Olam Social Poker EloGames14681494+28.3
BrokenArXiv 05/2026Math8.536.5+28.0
AA-Briefcase EloAgents11161589+27.5
ArXivMath 05/2026Math47.573.3+25.8
SWE-bench Verified Extended (Mercor held-out)Code30.054.0+24.0
WordleBenchGames97.073.0−24.0
AA-AnalystAgent (pass^5)Professional Work60.037.5−22.5
Terminal-Bench 4.0Code6.127.8+21.7
RuneBenchGames6.04.6−21.2
Legal Research BenchLegal34.655.3+20.7
Harvey's Legal Agent Benchmark - Antitrust CompetitionLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Arbitration International Dispute ResolutionLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Capital MarketsLegal20.040.0+20.0
Harvey's Legal Agent Benchmark - Corporate M&ALegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Energy Natural ResourcesLegal40.020.0−20.0
Harvey's Legal Agent Benchmark - Funds Asset ManagementLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Healthcare Life SciencesLegal20.040.0+20.0
Harvey's Legal Agent Benchmark - InsuranceLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - International Trade SanctionsLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Real EstateLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Structured Finance SecuritizationLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Trusts Estates Private ClientLegal0.020.0+20.0
TextQuests Game ProgressGames52.933.0−19.9
Terminal-Bench 4.0 (Artificial Analysis)Code13.633.3+19.7
LLM Debate BenchmarkChat & Writing14551593+19.5
Finance Agent v2 - General Qualitative AnalysisFinance & Accounting65.383.4+18.1
AA-LCR Extended (Mercor held-out)Search48.966.9+18.0
LMArena Agent Arena Net ImprovementAgents-1.33.8+16.9
DistressBenchSafety69.084.9+15.9
Tax Agent BenchFinance & Accounting57.773.4+15.7
Vibe Code Bench v1.1Code70.485.9+15.5
NoCoT-Bench NCRI 15.2General QA12698.3−15.3
𝜏³-BankingAgents35.550.5+15.1
HackerRank-BenchCode28.542.9+14.4
LLM Creative Story WritingChat & Writing-0.70.6+14.3
ArXivMath 06/2026Math62.576.7+14.2
Arena Code WebDev — Brand & MarketingCode15531644+13.3
ChartographyMultimodal40.427.6−12.8
Code MigrationCode34.847.4+12.6
GDPval-AA v2.1Professional Work14041681+12.2
DAYJOB: FinanceFinance & Accounting2.814.8+12.0
Mystery Game PuzzlesGames37.025.0−12.0
Shroombench (mushroom identification, top-1)Multimodal61.049.0−12.0
GeneBench-Pro Public Release SubsetScience14.026.0+12.0
CritPt (avg@5)Science14.326.0+11.7
Arena Code WebDev — ReactCode15841669+11.4
IOI (Vals v2)Code67.856.6−11.3
GDPval-AA v2Professional Work14331662+11.3
RiemannBenchMath39.228.0−11.2
ClockBench AccuracyMultimodal53.342.2−11.1
Arena Code WebDev — GamingCode16381731+10.4
Arena Code WebDev — FrontendCode15921672+10.3
DeepSWE v1.1Code65.575.4+9.9
FrontierMath Tier 4 (v2) - PrivateMath36.646.3+9.8
SpeciEval (Pooled English Runs)Safety89.098.7+9.7
SpeciEval (Refusal-Aware Retries, Historical)Safety89.098.8+9.7
Bug Hunt BenchCode21.030.7+9.7
ComplexConstraintsChat & Writing42.451.9+9.5
Creative Writing v3 EloChat & Writing17201907+9.5
Finance Agent v2 - Financial ModelingFinance & Accounting24.233.5+9.4
BuseyBench SVGMultimodal6.65.7−9.3
ArXivMath 04/2026Math57.566.7+9.2
Artificial Analysis Intelligence Index v4.3.2Indexes39.148.1+9.0
APEX-Agents 1.1Professional Work79.470.4−9.0
Artificial Analysis Intelligence Index v4.3Indexes39.448.2+8.7
Terminal-Bench Science 0.1 (Vals)Science5.714.3+8.6
Arena Code WebDev — Reference-Based DesignCode16081672+8.5
Arena Code WebDev — HTMLCode15931653+8.3
LMArena Agent Arena Net Improvement (Steering Burden)Agents-2.22.2+8.3
RSI-Exam 0.1 (Full 88)Agents0.30.4+8.2
Arena Code WebDev — SimulationsCode16181690+8.1
Arena Image-to-WebDevCode16001637+8.1
TaxCalcBench TY25Finance & Accounting14.06.0−8.0
ChessBench CoherenceGames83.291.2+8.0
DAYJOB: HealthcareHealth & Medicine0.07.8+7.8
APEX-Accounting 1.1Finance & Accounting57.264.9+7.7
LiveBench 2.0Agents67.274.8+7.6
HardGeoBench <=10miMultimodal46.338.8−7.5
HardGeoBench Country AccuracyMultimodal67.560.0−7.5
MedXpertQA MM (200-question set)Health & Medicine83.576.3−7.2
APEX-1 (Mercor, mean score)Professional Work67.073.9+6.9
Vals Index v2.1Indexes51.358.2+6.9
Gray Swan IPI Attack Success (k=15)Safety9.215.9−6.7
ArXivLean 06/2026Math21.728.3+6.5
Arena Code WebDev — FullstackCode16291653+6.3
Harvey HLAB — Criteria Pass RateLegal88.694.7+6.1
Arena Code WebDev — Content Creation ToolsCode15741612+6.0
EnterpriseBench: CoreCraftProfessional Work55.961.9+6.0
NoCoT-Bench NCKI (kspine v3)General QA121113−5.8
LMArena Agent Arena Bash RecoveryAgents-5.46.1+5.8
FrontierSWE V2Code20.325.8+5.5
Roboflow Vision Evals (September 2026 protocol)Multimodal85.279.8−5.4
Vals Index v2Indexes59.364.5+5.2
LMArena Agent Arena Steering BurdenAgents-7.32.6+5.0
HANDBOOK.mdProfessional Work10.815.4+4.6
Finance Agent v2 - Market AnalysisFinance & Accounting74.569.9−4.6
PCT Political Inconsistency (CAIS)Safety45.140.6+4.5
Arena Code WebDev — Consumer ProductCode15741609+4.3
AutomationBench-AA 1.0.6Agents62.057.9−4.2
Eval Connections One-Shot with TrapsGames80.084.0+4.0
HardGeoBench Avg DistanceMultimodal8951022−3.9
Excel Modeling BenchmarkFinance & Accounting71.367.4−3.9
Finance Agent v2 - General Quantitative AnalysisFinance & Accounting81.878.0−3.8
ChessBench EloGames12651201−3.8
GDP.pdfProfessional Work23.827.6+3.8
ALE-BenchCode18761981+3.7
LMArena Agent Arena Confirmed SuccessAgents-3.53.7+3.6
LMCAGeneral QA50.453.9+3.5
ChessBench AccuracyGames79.276.0−3.4
AIME 2025Math96.7100.0+3.3
Document ArenaProfessional Work14761473−3.2
Finance Agent v2 - ComparablesFinance & Accounting50.347.2−3.1
Terminal-Bench 2.1Code85.888.8+3.0
APEX-SWECode42.039.1−2.9
FrontierMath Tiers 1-3 (v2) - PrivateMath71.674.4+2.8
CharXiv Extended (Mercor held-out)Multimodal79.181.8+2.7
Arena Alignment IndexSafety74.477.1+2.6
MedXpertQA MM Extended (Mercor held-out)Health & Medicine62.760.1−2.6
MMMU-Pro Extended (Mercor held-out)Multimodal36.433.8−2.6
Humanity's Last Exam DiamondGeneral QA31.233.7+2.5
ProgramBenchCode0.02.5+2.5
Artificial Analysis Progress Index v1Indexes317324+2.5
Humanity's Last ExamGeneral QA45.643.2−2.4
Humanity's Sixth SenseMultimodal39.837.4−2.4
Finance Agent v2 - Disclosure AnalysisFinance & Accounting67.770.1+2.3
LMArena Agent Arena Praise vs ComplaintAgents-2.91.7+2.3
Conceptual Reasoning IndexIndexes64.666.2+2.2
Creative Writing v3 Rubric ScoreChat & Writing15.215.6+2.1
LMArena Agent Arena Bash ReliabilityAgents6.22.0−2.1
Finance Agent v2 - All PassFinance & Accounting47.549.5+2.1
OTIS Mock AIME 2024-2025Math97.299.2+1.9
Finance Agent v2 - AdjustmentsFinance & Accounting54.752.8−1.9
GDPval Extended (Mercor held-out rubric score)Professional Work57.855.9−1.9
Finance Agent v2 - Earnings AnalysisFinance & Accounting79.177.2−1.9
ACCoRDGeneral QA77.075.1−1.9
Arena Code WebDev — Data & AnalyticsCode15771588+1.7
Finance Agent v2 - PrecedentsFinance & Accounting33.835.2+1.4
CharXiv (descriptive and reasoning combined)Multimodal95.393.9−1.4
Vals RSI Index v1.1 (four tasks)Agents18.319.6+1.4
HardGeoBench <=100miMultimodal56.355.0−1.2
LMArena Text Arena EloChat & Writing14891496+1.2
MineBench — Bradley-Terry RatingGames17651747−1.1
LMArena Agent Arena SteerabilityAgents-0.01.6+0.8
AA-Omniscience IndexIndexes26.525.0−0.7
ObviousBenchGeneral QA100.099.3−0.7
Vision Arena OverallMultimodal12941291−0.6
GPQA Diamond (Artificial Analysis avg@5)General QA94.594.1−0.4
Epoch Capabilities Index (ECI)Indexes158158−0.4
ForecastBench Tournament Overall Brier IndexForecasting66.265.9−0.3
DTBenchGeneral QA96.896.5−0.3
ForecastBench Preliminary Dataset Brier IndexForecasting61.761.9+0.2
Ramp Accounting BenchFinance & Accounting42.342.5+0.2
GBENCH One-shot Coding GScoreGames52.652.7+0.1
SciCode 1.0.1 (Artificial Analysis)Code59.859.7−0.1
LMArena Agent Arena Tool HallucinationAgents0.70.80.0
AA-LCR v1.1Search83.083.00.0
CadQueryEval (Corrected Geometry)Code84.084.00.0
Harvey's Legal Agent Benchmark - Banking FinanceLegal20.020.00.0
Harvey's Legal Agent Benchmark - Employment LaborLegal0.00.00.0
Harvey's Legal Agent Benchmark - Environmental ESGLegal0.00.00.0
Harvey's Legal Agent Benchmark - Intellectual PropertyLegal0.00.00.0
Harvey's Legal Agent Benchmark - Litigation Dispute ResolutionLegal0.00.00.0
Harvey's Legal Agent Benchmark - TaxLegal0.00.00.0
Harvey's Legal Agent Benchmark - White Collar Defense InvestigationsLegal0.00.00.0
ProofBench v1.1Math58.058.00.0

Gemini 3.7 FlashvsMuse Spark 1.3normalised score marginsData Nov 15, 2026Data: Vals AI, Arena, Artificial Analysis and 37 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…