Atlas

Compare Models

Gemini 4 Argon vs Muse Spark 1.3

Gemini 4 Argon leads by 4.7 capability points (≈95% interval +2.5 to +7.0): a clear lead over Muse Spark 1.3 · P(higher) ≈ 100%

Wins 136 of 207 shared benchmarks (56 lost, 15 tied)

Capability vs. relative cost
Gemini 4 ArgonMuse 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 4 ArgonvsMuse Spark 1.3capability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins136 – 56 · 207 shared, 15 tied

Comparing matching setups only (33 benchmarks excluded) · suites counted once
← Gemini 4 ArgonMuse Spark 1.3 →

Gemini 4 ArgonvsMuse Spark 1.3wins by categoryData Nov 15, 2026Data: Vals AI, Artificial Analysis, Surge AI and 46 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 4 Argon.

CheatBenchSafety28.886.5−57.7
SAFE-Teen 106 worst branch: Health Advice (no prompt)Safety70.813.9+56.9
AA-PersonalityBench Imagination indexChat & Writing46.798.1+51.9
SAFE-Teen 106: Health Advice (no prompt)Safety57.46.7+50.7
RiemannBenchMath76.028.0−48.0
TextQuests Game ProgressGames77.333.0−44.4
ChartographyMultimodal71.627.6−44.0
IOI (Vals v2)Code100.056.6−43.4
TaxCalcBench TY25 (web search)Finance & Accounting74.032.0−42.0
ProofBench v1.1Math99.058.0−41.0
SRE Bench — Fully SolvedCode44.33.8−40.5
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 - Data Privacy CybersecurityLegal20.060.0+40.0
SAFE-Teen 106 worst branch: Health Advice (under-18 prompt)Safety43.64.5+39.1
Mystery Game PuzzlesGames64.025.0−39.0
Terminal-Bench-Science 0.1Science48.611.0−37.6
ArXivMath 08/2026 (Native Harnesses)Math82.545.6−36.8
Decision-Bench v1.1Professional Work54.519.8−34.7
Terminal-Bench Science 0.1 (Vals)Science44.310.0−34.3
BrokenArXiv 08/2026 (Native Harnesses)Math54.220.2−33.9
FrontierSWE V2 Worst@5Code44.512.6−31.9
FrontierMath Tier 4 (v2) - PrivateMath78.046.3−31.7
Terminal-Bench 2.1Code57.788.8+31.1
Terminal-Bench 4.0Code57.427.8−29.6
AutomationBench 1.0.6 (Zapier)Agents50.120.7−29.4
ALE-BenchCode28051981−29.2
NoCoT-Bench NCRI 15.2General QA15198.3−29.0
SAFE-Teen 106: Health Advice (under-18 prompt)Safety30.82.1+28.7
Humanity's Last Exam DiamondGeneral QA53.225.4−27.8
MineBench — Bradley-Terry RatingGames21951747−27.1
SAFE-Teen 106 worst branch: Urgent Safety (no prompt)Safety52.625.6+27.1
SAFE-Teen 106: Urgent Safety (no prompt)Safety33.19.0+24.1
FrontierML ($200 wallet)Agents39.516.0−23.6
FrontierSWE V2 Best@5Code64.340.9−23.4
Last Translation Benchmark (LTBv1-eval, blind)Chat & Writing40.918.4−22.5
Chess Puzzles (task version 2.0)Games59.037.0−22.0
Terminal-Bench 4.0 (Artificial Analysis)Code55.133.3−21.7
AutoResearchExam (hidden-test AUARC)Agents0.60.4−21.5
Code MigrationCode68.247.4−20.8
GameBench 2Games89.268.5−20.6
GDPval Extended (Mercor held-out rubric score)Professional Work76.455.9−20.5
Harvey's Legal Agent Benchmark - Arbitration International Dispute ResolutionLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Banking FinanceLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Funds Asset ManagementLegal40.020.0−20.0
Harvey's Legal Agent Benchmark - Healthcare Life SciencesLegal20.040.0+20.0
Harvey's Legal Agent Benchmark - ImmigrationLegal20.040.0+20.0
Harvey's Legal Agent Benchmark - Intellectual PropertyLegal20.00.0−20.0
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 ClientLegal40.020.0−20.0
HoneyBench v0.2 (hack rate)Safety63.683.6−20.0
AutomationBench-AA 1.0.6Agents77.557.9−19.7
Code2Video-AA EloCode12591019−19.5
AA-PersonalityBench Transparency indexChat & Writing21.941.0+19.2
SAFE-Teen 106 worst branch: Urgent Safety (under-18 prompt)Safety30.211.3+18.9
Ramp Accounting BenchFinance & Accounting61.142.5−18.6
BuseyBench SVGMultimodal7.65.7−18.5
Humanity's Sixth SenseMultimodal55.437.4−18.0
Visa Agentic Commerce BenchmarkAgents82.264.4−17.8
Legacy-Bench (Fireworks, avg@3)Code44.362.0+17.7
SpeciEval (Pooled English Runs)Safety81.398.7+17.5
AA-PersonalityBench Warmth indexChat & Writing79.196.2+17.3
AA-PersonalityBench Informativeness indexChat & Writing82.9100+17.3
DAYJOB: Data ScienceProfessional Work23.36.0−17.3
APEX-Agents 1.1Professional Work87.470.4−17.0
ChessBench EloGames14851201−16.9
Snorkel Agentic Coding 2.0Code41.825.0−16.8
AA-3D v1.0 EloCode11471032−16.7
DistressBenchSafety68.684.9+16.2
LMArena Agent Arena Net Improvement (Steering Burden)Agents10.72.2−16.1
GDP.xlsxProfessional Work38.322.6−15.7
Westworld Finance Diligence BenchFinance & Accounting60.745.2−15.6
GBENCH One-shot Coding GScoreGames67.952.7−15.2
Finance Agent v2 - PrecedentsFinance & Accounting49.835.2−14.6
Snorkel SWE-bench CLICode16.32.2−14.1
Artificial Analysis Intelligence Index v5.0.1Indexes50.436.6−13.8
LMArena Agent Arena Bash RecoveryAgents-21.46.1+13.7
AA-PersonalityBench Conviction indexChat & Writing63.877.2+13.5
AA-PersonalityBench Expressiveness indexChat & Writing84.871.4−13.5
CWE-bench v1 Programmatic Pass@1Cybersecurity68.055.0−13.0
SWE-bench Multimodal (Vals v1, 480 Tasks)Code56.043.1−12.9
DAYJOB: LawLegal9.922.8+12.9
Humanity's Last ExamGeneral QA55.643.2−12.4
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Legal27.415.0−12.4
Professional Reasoning Benchmark - LegalLegal49.461.6+12.2
Snorkel Digital Work IndexIndexes28.316.3−12.0
FrontierMath Tiers 1-3 (v2) - PrivateMath86.074.4−11.6
CursorBench 4.0Code52.941.6−11.3
AA-AnalystAgent (pass^5)Professional Work48.837.5−11.2
LMArena Agent Arena Praise vs ComplaintAgents24.11.7−11.2
ATLAS (Exa, 200-task subset)Search53.842.7−11.1
WellnessBenchSafety50.061.0+11.0
Vals RSI Index v1.1 (four tasks)Agents30.619.6−10.9
NoCoT-Bench NCKI (kspine v3)General QA128113−10.8
HackerRank-BenchCode49.738.9−10.8
SAFE-Teen 106 worst branch: Relationships (under-18 prompt)Safety14.03.2+10.8
LegalReasonBenchLegal50.840.4−10.4
RSI-Exam 0.1 (Full 88)Agents0.50.4−10.4
APEX-SWECode49.239.1−10.1
LMCAGeneral QA63.753.9−9.8
Finance Agent v2 - Market AnalysisFinance & Accounting79.769.9−9.8
Professional Reasoning Benchmark - FinanceFinance & Accounting49.959.5+9.6
SnorkelLegalLegal46.036.5−9.5
SAFE-Teen 106 worst branch: Relationships (no prompt)Safety65.756.2+9.5
MysteryMechanismScience45.536.0−9.5
SAFE-Teen 106: Urgent Safety (under-18 prompt)Safety12.12.7+9.4
LiveBench 2.0Agents84.174.8−9.3
SnorkelManufacturingProfessional Work19.510.5−9.0
Artificial Analysis Coding Agent Index v2.1Indexes45.536.8−8.7
AA-Omniscience IndexIndexes42.425.0−8.7
Conceptual Reasoning IndexIndexes72.466.2−8.6
Bug Hunt BenchCode38.730.7−8.1
Eval Connections One-Shot with TrapsGames92.084.0−8.0
Excel Modeling BenchmarkFinance & Accounting75.267.4−7.8
AA-LCR Extended (Mercor held-out)Search74.766.9−7.8
Finance Agent v2 - AdjustmentsFinance & Accounting60.552.8−7.7
AA-PersonalityBench Organization indexChat & Writing56.263.8+7.7
AA-PersonalityBench Slop indexChat & Writing67.675.3+7.7
Finance Agent v2 - Earnings AnalysisFinance & Accounting84.877.2−7.6
Creative Writing v3 EloChat & Writing17581907+7.5
Artificial Analysis Coding Agent Index v2.0Indexes53.245.7−7.5
Epoch Capabilities Index (ECI)Indexes167158−6.9
MedXpertQA MM Extended (Mercor held-out)Health & Medicine66.760.1−6.6
APEX-Accounting 1.1Finance & Accounting71.564.9−6.6
Finance Agent v2 - General Qualitative AnalysisFinance & Accounting77.083.4+6.4
Finance Agent v2 - All PassFinance & Accounting55.849.5−6.3
ChessBench AccuracyGames81.876.0−6.2
Artificial Analysis Progress Index v1Indexes342324−6.1
τ³-Banking (Mercor run, bm25_grep retrieval)Agents58.152.2−5.9
SAFE-Teen 106: Relationships (no prompt)Safety28.534.4−5.8
SnorkelRevOpsProfessional Work18.412.6−5.8
Tax Agent BenchFinance & Accounting79.073.4−5.6
micro1 PharmaBenchHealth & Medicine8.613.9+5.3
AA-Briefcase v1.1 EloAgents14851581+5.3
LMArena Agent Arena Steering BurdenAgents13.22.6−5.3
CyberBench v1.1 OverallCybersecurity77.972.7−5.1
SnorkelFinance 2.0Finance & Accounting15.110.0−5.1
Harvey LAB-AA v1.1 (hallucination-gated)Legal3.98.9+5.0
sudo L7Code18.913.9−5.0
HiL-Bench (Human-in-Loop Benchmark)Agents62.066.9+4.9
CritPt v2 (agentic, avg@5)Science72.968.0−4.9
LMArena Agent Arena Confirmed SuccessAgents13.23.7−4.8
LMArena Text Arena EloChat & Writing15221496−4.6
ErdosBenchMath3.02.9−4.5
Artificial Analysis Intelligence Index v4.3.2Indexes52.648.1−4.5
Finance Agent v2 - Disclosure AnalysisFinance & Accounting74.570.1−4.4
DeepFinanceFinance & Accounting41.437.0−4.4
HardGeoBench Avg DistanceMultimodal8801022−4.4
DAYJOB: FinanceFinance & Accounting18.814.8−4.0
Finance Agent v2 - ComparablesFinance & Accounting51.047.2−3.8
HardGeoBench <=10miMultimodal42.538.8−3.7
Finance Agent v2 - Financial ModelingFinance & Accounting29.833.5+3.7
SnorkelUnderwrite 2.0Finance & Accounting23.319.6−3.7
Access to Justice BenchmarkLegal68.572.2+3.7
LMArena Agent Arena Bash ReliabilityAgents-5.42.0+3.7
LMArena Code Arena WebDevCode16821655−3.5
Finance Agent v2 - General Quantitative AnalysisFinance & Accounting81.578.0−3.5
AA-LCR v1.1Search79.783.0+3.3
DrugDiscoveryBenchScience65.462.2−3.2
DAXBench v2Code88.385.1−3.2
Immigration BenchLegal68.165.0−3.1
DAYJOB: HealthcareHealth & Medicine10.87.8−3.0
FrontierCode 1.1 Main (Artificial Analysis)Code43.540.6−2.8
LLM Creative Story WritingChat & Writing0.80.6−2.7
HANDBOOK.md Agents v1.1Professional Work36.633.9−2.7
FrontierCode 2.0 (Artificial Analysis)Code25.723.1−2.6
DeepSWE v1.1Code77.975.4−2.5
HardGeoBench Country AccuracyMultimodal62.560.0−2.5
SWE-Marathon v1.1Code43.846.3+2.5
SAFE-Teen 106: Relationships (under-18 prompt)Safety3.00.6+2.4
Arena Alignment IndexSafety79.477.1−2.4
ACCoRDGeneral QA77.375.1−2.2
Financial Coaching BenchFinance & Accounting73.171.0−2.1
GDPval-AA v2.1Professional Work16341681+2.1
SciCode 1.0.1 (Artificial Analysis)Code61.859.7−2.1
PCT Political Inconsistency (CAIS)Safety38.740.6−1.9
Longform Creative WritingChat & Writing80.982.8+1.9
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine13.915.6+1.7
ChessBench CoherenceGames92.691.2−1.4
ComplexConstraintsChat & Writing53.351.9−1.4
CritPt (avg@5)Science27.126.0−1.1
DTBenchGeneral QA97.696.5−1.1
CUA-benchAgents4.85.8+1.0
OTIS Mock AIME 2024-2025Math100.099.2−0.8
GDP.pdfProfessional Work28.427.6−0.8
Vibe Code Bench 1–100Code21.220.5−0.7
Harvey HLAB — Criteria Pass RateLegal94.094.7+0.7
Legal Research BenchLegal54.855.3+0.5
Creative Writing v3 Rubric ScoreChat & Writing15.615.6+0.4
DAYJOB: TaxFinance & Accounting3.23.6+0.4
LMArena Agent Arena Tool HallucinationAgents0.50.8+0.1
DAYJOB: AccountingFinance & Accounting18.018.00.0
HardGeoBench <=100miMultimodal55.055.00.0
Harvey's Legal Agent Benchmark - Antitrust CompetitionLegal20.020.00.0
Harvey's Legal Agent Benchmark - Capital MarketsLegal40.040.00.0
Harvey's Legal Agent Benchmark - Corporate M&ALegal20.020.00.0
Harvey's Legal Agent Benchmark - Emerging Companies Venture CapitalLegal40.040.00.0
Harvey's Legal Agent Benchmark - Employment LaborLegal0.00.00.0
Harvey's Legal Agent Benchmark - Energy Natural ResourcesLegal20.020.00.0
Harvey's Legal Agent Benchmark - Environmental ESGLegal0.00.00.0
Harvey's Legal Agent Benchmark - InsuranceLegal0.00.00.0
Harvey's Legal Agent Benchmark - International Trade SanctionsLegal20.020.00.0
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
ProgramBenchCode2.52.50.0

Gemini 4 ArgonvsMuse Spark 1.3normalised score marginsData Nov 15, 2026Data: Vals AI, Artificial Analysis, Surge AI and 46 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…