Atlas

Compare Models

Gemini 4 Argon vs Muse Wonder

Gemini 4 Argon leads by 2.4 capability points (≈95% interval −0.0 to +4.8): statistically tied · P(higher) ≈ 97%

Wins 139 of 221 shared benchmarks (66 lost, 16 tied)

Capability vs. relative cost
Gemini 4 ArgonMuse 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.

Gemini 4 ArgonvsMuse Wondercapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins139 – 66 · 221 shared, 16 tied

Comparing matching setups only (27 benchmarks excluded) · suites counted once
← Gemini 4 ArgonMuse Wonder →

Gemini 4 ArgonvsMuse Wonderwins by categoryData Nov 15, 2026Data: Vals AI, Artificial Analysis, Center for AI Safety and 55 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Muse Wonder minus Gemini 4 Argon.

HPCT Refusal (CAIS)Safety29.097.0+68.0
LisanBench Path LengthGames756027973+63.4
VCT-Refusal (CAIS)Safety10.760.2+49.5
LisanBench Difficulty-WeightedGames38498303+47.7
CheatBenchSafety28.873.1−44.3
AA-PersonalityBench Imagination indexChat & Writing46.790.5+44.2
Harvey's Legal Agent Benchmark - Emerging Companies Venture CapitalLegal40.00.0−40.0
Harvey's Legal Agent Benchmark - Litigation Dispute ResolutionLegal0.040.0+40.0
AA-PersonalityBench Organization indexChat & Writing56.290.5+34.6
AA-PersonalityBench Transparency indexChat & Writing21.956.2+34.6
EBR-bench v4 (card ban, single agent)Games64.831.4−33.3
IOI (Vals v2)Code100.066.9−33.1
TextQuests Game ProgressGames77.347.6−29.7
NoCoT-Bench NCRI 15.2General QA15197.7−29.3
EnigmaEvalGeneral QA64.636.1−28.5
TaxCalcBench TY25 (web search)Finance & Accounting74.046.0−28.0
DAYJOB: LawLegal9.936.6+26.7
ChessBench EloGames14851046−26.1
MASKSafety61.084.9+23.9
EQ-Bench 4Chat & Writing12361325+23.2
Terminal-Bench-Science 0.1Science48.625.7−22.9
CrossSearch (pass^3)Search64.042.0−22.0
Terminal-Bench Science 0.1 (Vals)Science44.322.9−21.4
Code2Video-AA EloCode1259997−21.3
AA-PersonalityBench Warmth indexChat & Writing79.1100+21.2
FrontierML ($200 wallet)Agents39.519.3−20.3
HardGeoBench <=100miMultimodal58.838.8−20.0
Harvey's Legal Agent Benchmark - Banking FinanceLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Capital MarketsLegal40.020.0−20.0
Harvey's Legal Agent Benchmark - Funds Asset ManagementLegal40.020.0−20.0
Harvey's Legal Agent Benchmark - ImmigrationLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - International Trade SanctionsLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - Real EstateLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Trusts Estates Private ClientLegal40.020.0−20.0
Decision-Bench v1.1Professional Work54.534.6−20.0
MineBench — Bradley-Terry RatingGames21951868−19.8
HardGeoBench Avg DistanceMultimodal7411369−19.4
Eval Connections One-Shot with TrapsGames92.073.0−19.0
Furniture Assembly BenchmarkMultimodal72.554.2−18.3
DataCenterBench Tier 1Agents35.016.7−18.3
Build-A-Chip-Bench (20-task index run)Agents18.60.3−18.3
TextQuests Average Cumulative HarmSafety26.120.8+18.2
MysteryMechanismScience45.527.5−18.0
Finance Agent v2 - PrecedentsFinance & Accounting49.832.1−17.7
HiL-Bench (Human-in-Loop Benchmark)Agents62.079.7+17.7
Finance Agent v2 - Market AnalysisFinance & Accounting79.762.1−17.6
HardGeoBench <=10miMultimodal45.027.5−17.5
HoneyBench v0.2 (hack rate)Safety63.681.0−17.4
DistressBenchSafety68.685.9+17.3
LMArena Agent Arena Net Improvement (Steering Burden)Agents10.71.8−16.9
Terminal-Bench 4.0 (Artificial Analysis)Code55.138.4−16.7
SpeciEval (Pooled English Runs)Safety81.397.5+16.3
Vals Science IndexIndexes55.439.1−16.2
Humanity's Last Exam DiamondGeneral QA53.237.3−15.9
VIALSMultimodal47.231.5−15.7
AA-PersonalityBench Conviction indexChat & Writing63.879.1+15.4
RiemannBench (November 2026 rescoring)Math74.459.2−15.2
ProofBench v1.1Math99.084.0−15.0
AA-PersonalityBench Slop indexChat & Writing67.654.3−13.5
FrontierMemory v1.1Agents64.451.1−13.3
NoCoT-Bench NCKI (kspine v3)General QA128109−13.1
LLM Creative Story WritingChat & Writing0.82.0+13.1
Mystery Game PuzzlesGames64.051.0−13.0
LMArena Agent Arena Praise vs ComplaintAgents24.1-1.8−13.0
DAYJOB: AccountingFinance & Accounting18.030.8+12.8
Creative Writing v3 EloChat & Writing17582007+12.6
ChessBench AccuracyGames81.870.2−12.4
Ramp Accounting BenchFinance & Accounting61.148.7−12.4
RWE-benchAgents12.00.0−12.0
Humanity's Sixth SenseMultimodal55.443.7−11.7
BioTIER RefusalSafety87.298.7+11.5
AA-PersonalityBench Informativeness indexChat & Writing82.994.3+11.5
GameBench 2Games89.277.8−11.3
OpenWorldEngineer (Mean@8)Code24.112.7−11.3
Dual-Arm Sim (task completion)Agents11.30.0−11.2
ERQAMultimodal86.575.3−11.2
HardGeoBench Country AccuracyMultimodal63.852.5−11.2
SpatialViz-BenchMultimodal94.283.5−10.8
SWE-bench Multimodal (Vals v1, 480 Tasks)Code56.045.6−10.4
APEX-Agents 1.1Professional Work87.477.0−10.4
FrontierBioMLScience36.826.9−10.0
Finance Agent v2 - All PassFinance & Accounting55.845.9−9.9
GBENCH One-shot Coding GScoreGames67.958.1−9.8
BuseyBench SVGMultimodal7.66.6−9.6
AutomationBench-AA 1.0.6Agents77.568.1−9.4
OncologyBench Strict (25 tasks, October 9 rubric)Health & Medicine33.324.0−9.3
BioMysteryBench (Vals Terminus 2)Science76.367.0−9.3
BullshitBench v2 LeaderboardSafety66.057.0−9.0
Professional Reasoning Benchmark - LegalLegal49.458.4+9.0
Professional Reasoning Benchmark - FinanceFinance & Accounting49.958.7+8.8
SWE-Marathon v1.1Code43.852.5+8.8
micro1 PharmaBenchHealth & Medicine8.617.2+8.6
Bug Hunt BenchCode38.730.2−8.6
AA-Omniscience IndexIndexes42.425.5−8.4
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine13.922.2+8.3
SnorkelManufacturingProfessional Work19.527.8+8.3
Excel Modeling BenchmarkFinance & Accounting75.267.1−8.2
Terminal-Bench 4.0Code57.649.5−8.1
Finance Agent v2 - Earnings AnalysisFinance & Accounting84.876.8−8.0
Humanity's Last ExamGeneral QA55.647.6−8.0
DualEntry Accounting AI BenchmarkFinance & Accounting91.183.2−7.9
Tax Agent BenchFinance & Accounting79.071.2−7.9
Finance Agent v2 - AdjustmentsFinance & Accounting60.552.7−7.9
SnorkelLegalLegal46.038.3−7.7
ChessBench CoherenceGames92.6100.0+7.4
FrontierMath Tier 4 (v2) - PrivateMath78.070.7−7.3
Toloka Arena (v3)Agents73.065.7−7.3
Finance Agent v2 - Financial ModelingFinance & Accounting29.822.5−7.3
Finance Agent v2 - Disclosure AnalysisFinance & Accounting74.567.3−7.2
DAYJOB: HealthcareHealth & Medicine10.818.0+7.2
HANDBOOK.md Agents v1.1Professional Work36.629.5−7.1
LMCAGeneral QA63.756.8−6.9
AA-Briefcase v1.1 EloAgents14851610+6.8
Artificial Analysis Intelligence Index v5.0.1Indexes50.443.6−6.8
DAYJOB: Data ScienceProfessional Work23.316.7−6.6
BioSecBench-Refusal V1 Clean RerunsSafety48.755.3+6.6
Conceptual Reasoning IndexIndexes72.467.7−6.4
GDPval Extended (Mercor held-out rubric score)Professional Work76.470.0−6.4
τ³-Banking (Mercor run, bm25_grep retrieval)Agents58.151.9−6.2
Finance Agent v2 - General Qualitative AnalysisFinance & Accounting77.070.8−6.2
GDPval-AA v2.1Professional Work16341770+6.0
LiveBench 2.0Agents84.178.1−6.0
STEMBench120 v1.0.0 (closed-book)Science2.42.7+6.0
MindCube TinyMultimodal96.791.1−5.7
Code MigrationCode68.262.6−5.6
PCT Political Inconsistency (CAIS)Safety38.733.3+5.4
APEX-SWECode49.254.6+5.4
Vibe Code Bench 1–100Code21.226.6+5.4
LMArena Agent Arena Steering BurdenAgents13.22.6−5.3
SWE Atlas v1.1 - OverallCode70.164.8−5.3
LMArena Agent Arena Bash ReliabilityAgents-5.44.8+5.1
WellnessBenchSafety50.055.0+5.0
GDP.pdf (Artificial Analysis, All-pass)Professional Work25.630.6+5.0
Snorkel SWE-bench CLICode16.311.4−4.9
Vals Finance IndexIndexes66.061.2−4.8
Access to Justice BenchmarkLegal68.573.0+4.5
IntPhys 2Multimodal74.670.2−4.5
APEX-Accounting 1.1Finance & Accounting71.567.1−4.4
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Legal27.423.0−4.4
LMArena Code Arena WebDevCode16821649−4.4
Legal Research BenchLegal54.850.5−4.3
Machiavelli Deception (CAIS)Safety85.083.6+4.3
DAXBench v2Code88.384.0−4.3
LMArena Agent Arena Confirmed SuccessAgents13.25.4−3.9
Finance Agent v2 - General Quantitative AnalysisFinance & Accounting81.577.6−3.9
FrontierMath Tiers 1-3 (v2) - PrivateMath86.082.1−3.9
AA-PersonalityBench Expressiveness indexChat & Writing84.888.6+3.8
Snorkel WorkplaceAgentsProfessional Work21.217.4−3.8
MKA (Artificial Analysis)Professional Work51.347.5−3.8
Epoch Capabilities Index (ECI)Indexes167162−3.7
MMMU ProMultimodal88.384.9−3.4
HLAB-Vals-1.1Legal27.123.8−3.3
sudo L7Code18.915.6−3.3
Creative Writing v3 Rubric ScoreChat & Writing15.616.2+3.3
Real-SWE 1.1Code39.242.4+3.2
Finance Agent v2 - ComparablesFinance & Accounting51.047.9−3.0
FrontierCode 1.1 Main (Artificial Analysis)Code43.546.5+3.0
MacroscopeBenchCode61.364.2+2.9
SpatialBench V1 Clean RerunsScience69.071.9+2.9
DeepFinanceFinance & Accounting41.444.2+2.8
DAYJOB: TaxFinance & Accounting3.26.0+2.8
CursorBench 4.0Code52.950.4−2.5
CyberBench v1.1 OverallCybersecurity77.980.4+2.5
DAYJOB: FinanceFinance & Accounting18.821.3+2.5
ComplexConstraints (November 2026 rescoring)Chat & Writing70.367.9−2.4
DTBenchGeneral QA97.695.2−2.4
CUA-benchAgents4.87.2+2.3
Excel Modeling Benchmark v1.1Finance & Accounting74.872.5−2.3
FrontierCode 2.0 - ScoreCode27.625.3−2.3
Finance Agent v2.1Finance & Accounting67.765.5−2.1
Artificial Analysis Progress Index v1Indexes342336−2.1
BioResearchBenchScience13.011.0−2.0
CritPt v2 (agentic, avg@5)Science72.970.9−2.0
MissionReportProfessional Work64.162.1−2.0
LMArena Text Arena EloChat & Writing15221511−2.0
Snorkel Digital Work IndexIndexes28.326.4−1.9
FrontierCode 2.0 - Pass RateCode30.628.8−1.8
Snorkel Agentic Coding 2.0Code41.843.6+1.8
Agents' Last Exam v1.1 (ALE v1.1)Agents40.238.5−1.6
Code Migration v1.1Code70.872.4+1.6
Immigration BenchLegal68.169.6+1.5
Vals Legal IndexIndexes76.875.3−1.5
SciCode 1.0.1 (Artificial Analysis)Code61.860.3−1.5
Harvey HLAB — Criteria Pass RateLegal94.092.7−1.3
Artificial Analysis Coding Agent Index v2.0Indexes53.252.0−1.3
CritPt (avg@5)Science27.128.3+1.1
ACCoRDGeneral QA77.376.2−1.1
CWE-bench v2 Pass@1Cybersecurity65.066.1+1.1
GDPval (Mercor 213-task rubric score)Professional Work85.987.0+1.1
Artificial Analysis Coding Agent Index v2.1Indexes45.544.5−1.0
DeepSWE v1.1Code77.978.8+0.9
Prophet Arena Arena Score (Agentic, OpenCode)Forecasting86.085.4−0.7
Financial Coaching BenchFinance & Accounting73.173.8+0.7
OTIS Mock AIME 2024-2025Math100.099.4−0.6
GDPxProfessional Work7.36.8−0.5
LMArena Agent Arena Tool HallucinationAgents0.5-0.5−0.5
Public Benefits Bench v1.1Professional Work69.870.2+0.5
SnorkelRevOpsProfessional Work18.418.0−0.4
HackerRank-BenchCode49.750.0+0.3
SnorkelUnderwrite 2.0Finance & Accounting23.323.0−0.3
AA-LCR Extended (Mercor held-out)Search74.774.4−0.3
ErdosBenchMath3.03.0+0.3
Artificial Analysis Intelligence Index v4.3.2Indexes52.652.4−0.2
Prophet Arena Arena Score (Fixed Context)Forecasting85.385.2−0.1
SnorkelFinance 2.0Finance & Accounting15.115.0−0.1
Harvey's Legal Agent Benchmark - Antitrust CompetitionLegal20.020.00.0
Harvey's Legal Agent Benchmark - Arbitration International Dispute ResolutionLegal0.00.00.0
Harvey's Legal Agent Benchmark - Bankruptcy RestructuringLegal0.00.00.0
Harvey's Legal Agent Benchmark - Corporate GovernanceLegal20.020.00.0
Harvey's Legal Agent Benchmark - Corporate M&ALegal20.020.00.0
Harvey's Legal Agent Benchmark - Data Privacy CybersecurityLegal20.020.00.0
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 - Healthcare Life SciencesLegal20.020.00.0
Harvey's Legal Agent Benchmark - InsuranceLegal0.00.00.0
Harvey's Legal Agent Benchmark - Intellectual PropertyLegal20.020.00.0
Harvey's Legal Agent Benchmark - Structured Finance SecuritizationLegal0.00.00.0
Harvey's Legal Agent Benchmark - TaxLegal0.00.00.0
Harvey's Legal Agent Benchmark - White Collar Defense InvestigationsLegal0.00.00.0
Visa Agentic Commerce BenchmarkAgents82.282.20.0

Gemini 4 ArgonvsMuse Wondernormalised score marginsData Nov 15, 2026Data: Vals AI, Artificial Analysis, Center for AI Safety and 55 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…