Atlas

Compare Models

Gemini 4 Argon vs Grok 4.6

Gemini 4 Argon leads by 5.9 capability points (≈95% interval +3.7 to +8.2): a clear lead over Grok 4.6 · P(higher) ≈ 100%

Wins 178 of 218 shared benchmarks (26 lost, 14 tied)

Capability vs. relative cost
Gemini 4 ArgonGrok 4.6Closed 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 ArgonvsGrok 4.6capability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins178 – 26 · 218 shared, 14 tied

Comparing matching setups only (37 benchmarks excluded) · suites counted once
← Gemini 4 ArgonGrok 4.6 →

Gemini 4 ArgonvsGrok 4.6wins by categoryData Nov 15, 2026Data: Vals AI, Artificial Analysis, benchmarks.bio and 50 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Grok 4.6 minus Gemini 4 Argon.

ARC-AGI-3 (Semi-Private)General QA97.52.1−95.3
IOI (Vals v2)Code100.047.6−52.4
ChartographyMultimodal71.620.5−51.1
CheatBenchSafety28.879.6−50.8
BrowseComp Extended (Mercor held-out)Search72.021.5−50.5
AA-PersonalityBench Transparency indexChat & Writing21.971.4+50.0
HieroglyphBenchMultimodal69.720.3−49.4
AA-PersonalityBench Warmth indexChat & Writing79.131.5−48.1
ProofBench v1.1Math99.051.0−48.0
LLM Chess EloGames1609443−47.9
Time Horizon Index: KSPAgents52.55.8−46.7
FrontierMath Tier 4 (v2) - PrivateMath78.031.7−46.3
Terminal-Bench-Science 0.1Science48.66.2−42.4
AA-PersonalityBench Organization indexChat & Writing56.214.3−42.3
Terminal-Bench 4.0 (Artificial Analysis)Code55.113.1−41.9
LLM Creative Story WritingChat & Writing0.8-2.9−41.6
Terminal-Bench 4.0Code57.617.2−40.4
Harvey's Legal Agent Benchmark - Capital MarketsLegal40.00.0−40.0
Harvey's Legal Agent Benchmark - Emerging Companies Venture CapitalLegal40.00.0−40.0
Harvey's Legal Agent Benchmark - Trusts Estates Private ClientLegal40.00.0−40.0
RareDiseaseBench V1 (35 evaluations)Science43.84.8−39.1
AA-PersonalityBench Informativeness indexChat & Writing82.944.8−38.5
NoCoT-Bench NCRI 15.2General QA15186.8−35.4
ChessBench EloGames1485891−35.3
ALE-BenchCode28051822−34.8
Humanity's Last Exam DiamondGeneral QA53.219.6−33.6
HardGeoBench Avg DistanceMultimodal7411807−32.9
Terminal-Bench Science 0.1 (Vals)Science44.311.4−32.9
AA-PersonalityBench Slop indexChat & Writing67.635.3−32.7
Furniture Assembly BenchmarkMultimodal72.540.0−32.5
Concept Synth INDUCTION Challenge100 Round 1General QA57.025.0−32.0
FrontierSWE V2 Worst@5Code44.512.9−31.6
Wiz Cyber Model Arena v3.9Cybersecurity77.246.6−30.6
Mystery Game PuzzlesGames64.034.0−30.0
Concept Synth INDUCTION Challenge64 Round 1General QA65.635.9−29.7
BuseyBench SVGMultimodal7.64.6−29.5
CrossSearch (pass^3)Search64.035.0−29.0
AA-PersonalityBench Conviction indexChat & Writing63.892.4+28.8
Finance Agent v2 - PrecedentsFinance & Accounting49.821.9−27.9
LAB-Bench ProtocolQA (SecureBio)Science95.968.1−27.8
FrontierSWE V2 Best@5Code64.337.3−27.0
AA-PersonalityBench Imagination indexChat & Writing46.720.0−26.9
SimpleQA Verified (Epoch anti-abstention)General QA76.049.3−26.7
VIALSMultimodal47.220.5−26.7
ChessBenchGames70.043.3−26.7
Bio Capabilities Index (SecureBio)Indexes174137−26.5
HardGeoBench <=100miMultimodal58.832.5−26.2
Chess Puzzles (task version 2.0)Games59.033.0−26.0
HardGeoBench <=10miMultimodal45.020.0−25.0
SAGEMath53.628.9−24.7
ARC-AGI-2General QA91.367.1−24.2
Code MigrationCode68.244.6−23.6
ChessBench CoherenceGames92.669.2−23.4
MineBench — Bradley-Terry RatingGames21951811−23.3
HackerRank-BenchCode49.727.9−21.8
Blueprint-Bench 2Multimodal0.50.3−21.2
AA-PersonalityBench Expressiveness indexChat & Writing84.863.8−21.2
LMArena Agent Arena Net Improvement (Steering Burden)Agents10.7-0.2−20.6
Terminal-Bench 2.1Code57.778.3+20.6
GameBench 2Games89.268.7−20.4
FrontierMath Tiers 1-3 (v2) - PrivateMath86.066.0−20.0
Harvey's Legal Agent Benchmark - Banking FinanceLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Energy Natural ResourcesLegal20.040.0+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 - Intellectual PropertyLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - International Trade SanctionsLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - Litigation Dispute ResolutionLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Structured Finance SecuritizationLegal0.020.0+20.0
HardGeoBench Country AccuracyMultimodal63.843.8−20.0
Artificial Analysis Intelligence Index v5.0.1Indexes50.431.0−19.4
TissueMicroscopyBench V1 (49 evaluations)Science21.82.7−19.0
MedXpertQA MM Extended (Mercor held-out)Health & Medicine66.748.4−18.3
CloudOpsBench v1.0Agents57.539.7−17.9
MirrorCode (ML, +Private, 2L)Code58.941.1−17.8
Finance Agent v2 - Market AnalysisFinance & Accounting79.762.2−17.5
Vals Science IndexIndexes55.438.1−17.3
Toloka Arena (v3)Agents73.056.0−17.0
ComplexConstraintsChat & Writing53.336.5−16.8
HealthBench ProfessionalHealth & Medicine64.248.5−15.7
LMCAGeneral QA63.748.5−15.2
ChessBench AccuracyGames81.867.8−15.0
MysteryMechanismScience45.530.6−14.9
OncologyBench Strict (25 tasks, October 9 rubric)Health & Medicine33.318.7−14.7
GDPval Extended (Mercor held-out rubric score)Professional Work76.461.8−14.6
Humanity's Last Exam (Text-Only)General QA57.142.9−14.2
MedCodeHealth & Medicine58.844.7−14.1
Conceptual Reasoning IndexIndexes72.462.5−13.6
LMArena Agent Arena Bash RecoveryAgents-21.45.7+13.5
LMArena Agent Arena Praise vs ComplaintAgents24.1-2.5−13.3
Finance Agent v2 - All PassFinance & Accounting55.842.9−12.9
MBCT (SecureBio)Science63.250.4−12.8
RSI-Exam 0.1 (Full 88)Agents0.50.4−12.7
DeepSWE v1.1Code77.965.2−12.7
HPCT (SecureBio)Science74.261.6−12.6
GBENCH One-shot Coding GScoreGames67.955.4−12.5
Excel Modeling BenchmarkFinance & Accounting75.262.7−12.5
τ³-Banking (Mercor run, bm25_grep retrieval)Agents58.145.7−12.4
BioSecBench-Surveillance V1 Clean RerunsScience59.847.5−12.3
ScBench-Long V1 Clean RerunsScience28.416.5−11.9
SWE-Marathon v1.1Code43.831.9−11.9
NoCoT-Bench NCKI (kspine v3)General QA128112−11.5
LMArena Text Arena EloChat & Writing15221456−11.5
CursorBench 4.0Code52.941.4−11.5
Bug Hunt BenchCode38.727.3−11.4
APEX-Accounting 1.1Finance & Accounting71.560.3−11.2
GDP.pdfProfessional Work28.417.2−11.2
CWE-bench v1 Programmatic Pass@1Cybersecurity68.057.0−11.0
ARC-AGI-1 Semi-PrivateGeneral QA98.587.5−11.0
Finance Agent v2 - Earnings AnalysisFinance & Accounting84.873.9−10.9
Finance Agent v2 - AdjustmentsFinance & Accounting60.549.6−10.9
Finance Agent v2 - Financial ModelingFinance & Accounting29.818.9−10.9
WCB (SecureBio)Science61.150.3−10.8
TxBench Antibody Discovery V1 (100 evaluations)Science63.753.0−10.7
Snorkel SWE-bench CLICode16.35.7−10.6
CritPt v2 (agentic, avg@5)Science72.962.3−10.6
AutomationBench-AA 1.0.6Agents77.567.0−10.5
CyberBench v1.1 OverallCybersecurity77.967.7−10.2
Finance Agent v2 - Disclosure AnalysisFinance & Accounting74.564.4−10.1
AutoResearchExam (hidden-test AUARC)Agents0.60.5−10.1
APEX-Agents 1.1Professional Work87.477.3−10.1
VCT (SecureBio)Science58.448.3−10.1
SnorkelFinance 2.0Finance & Accounting15.125.1+10.0
VariantBench V1 Clean RerunsScience50.340.4−9.9
Finance Agent v2 - ComparablesFinance & Accounting51.041.2−9.7
SpeciEval (Pooled English Runs)Safety81.390.8+9.6
AA-LCR Extended (Mercor held-out)Search74.765.2−9.5
Bedside BenchHealth & Medicine92.783.3−9.4
LiveBench 2.0Agents84.174.9−9.2
LMArena Code Arena WebDevCode16821618−8.4
Finance Agent v2 - General Quantitative AnalysisFinance & Accounting81.573.2−8.3
Artificial Analysis Intelligence Index v4.3.2Indexes52.644.3−8.2
Snorkel Agentic Coding 2.0Code41.833.6−8.2
DAYJOB: HealthcareHealth & Medicine10.82.8−8.0
LMArena Agent Arena Confirmed SuccessAgents13.2-2.6−7.9
MMMU ProMultimodal88.380.5−7.8
Snorkel Digital Work IndexIndexes28.320.5−7.8
LMArena Agent Arena Bash ReliabilityAgents-5.410.0+7.7
Tax Agent BenchFinance & Accounting79.071.4−7.6
τ³-Banking (Fireworks, banking_knowledge)Agents58.250.7−7.6
AA-AnalystAgent (pass^5)Professional Work48.841.3−7.5
LMArena Agent Arena Steering BurdenAgents13.2-1.8−7.5
CritPt (avg@5)Science27.119.7−7.4
APEX-SWECode49.256.4+7.2
Artificial Analysis Progress Index v1Indexes342322−7.1
WellnessBenchSafety50.043.0−7.0
Epoch Capabilities Index (ECI)Indexes167158−7.0
Legal Research BenchLegal54.848.1−6.7
Vibe Code Bench 1–100Code21.214.8−6.4
Design Arena Overall Code CategoriesCode13261288−6.3
BioSecBench-RespiratoryVirus-Surveillance V1 (110 evaluations)Science17.611.5−6.1
DAYJOB: FinanceFinance & Accounting18.812.8−6.0
FrontierCode 2.0 (Artificial Analysis)Code25.719.8−6.0
AA-Omniscience IndexIndexes42.430.5−5.9
LAB-Bench CloningScenarios (SecureBio)Science100.094.2−5.8
ACCoRDGeneral QA77.371.6−5.7
SnorkelUnderwrite 2.0Finance & Accounting23.317.8−5.5
EpiBench V1 Clean RerunsScience28.923.6−5.4
SciCode 1.0.1 (Artificial Analysis)Code61.856.5−5.3
SnorkelLegalLegal46.040.7−5.3
Vals RSI Index v1.1 (four tasks)Agents30.625.5−5.1
TxBench Oligo Discovery V1 (113 evaluations)Science51.046.0−5.0
TxBench-Antibody-Development V1 (60 evaluations)Science31.126.1−5.0
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine13.918.3+4.4
Snorkel WorkplaceAgentsProfessional Work21.217.1−4.1
BioMysteryBench (Vals Terminus 2)Science76.372.2−4.1
micro1 PharmaBenchHealth & Medicine8.612.5+3.9
MetagenomicsBench V1Science42.338.7−3.7
SnorkelManufacturingProfessional Work19.516.4−3.1
TxBench Preclinical Pharmacology V1 Clean RerunsScience64.761.7−3.0
Public Benefits Bench v1.1Professional Work69.866.8−2.9
Access to Justice BenchmarkLegal68.571.3+2.9
SnorkelRevOpsProfessional Work18.415.8−2.6
GPQA DiamondGeneral QA96.494.0−2.4
SpatialBench-Long V1 Clean RerunsScience28.626.2−2.4
MacroscopeBenchCode72.875.1+2.3
AA-Briefcase v1.1 EloAgents14851526+2.3
DAXBench v2Code88.386.2−2.1
Immigration BenchLegal68.166.0−2.1
LegalBenchLegal88.386.3−2.0
GPQA-Bio (SecureBio)Science89.387.4−1.9
BioSecBench-Refusal V1 Clean RerunsSafety48.746.9−1.8
SpatialBench V1 Clean RerunsScience69.067.3−1.7
Harvey HLAB — Criteria Pass RateLegal94.092.5−1.5
ProgramBenchCode2.51.0−1.5
NeuroscienceBench V1 (91 evaluations)Science19.120.5+1.4
Senior SWE-BenchCode27.726.3−1.4
AA-LCR v1.1Search79.781.0+1.3
Arena Alignment IndexSafety79.480.8+1.3
Eval Connections One-Shot with TrapsGames92.091.0−1.0
FrontierCode 1.1 Extended - Pass RateCode66.167.0+0.9
MedScribeHealth & Medicine87.486.5−0.9
OTIS Mock AIME 2024-2025Math100.099.2−0.8
Finance Agent v2 - General Qualitative AnalysisFinance & Accounting77.077.8+0.8
FrontierCode 1.1 Extended - ScoreCode60.661.3+0.7
BioSecBench-Structure V1 (58 evaluations)Science14.915.5+0.6
Prophet Arena Arena Score (Agentic, OpenCode)Forecasting86.085.6−0.5
BioSecBench-Function V1 Clean RerunsScience44.444.0−0.4
Prophet Arena Arena Score (Fixed Context)Forecasting85.385.0−0.3
MMLU-Bio (SecureBio)Science92.191.8−0.3
GDPval-AA v2.1Professional Work16341627−0.3
DTBenchGeneral QA97.697.3−0.3
LMArena Agent Arena Tool HallucinationAgents0.51.0+0.2
GDPval (Mercor 213-task rubric score)Professional Work85.986.0+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 - 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 - Real EstateLegal0.00.00.0
Harvey's Legal Agent Benchmark - TaxLegal0.00.00.0
Harvey's Legal Agent Benchmark - White Collar Defense InvestigationsLegal0.00.00.0
ScBench V1 Clean RerunsScience59.159.10.0

Gemini 4 ArgonvsGrok 4.6normalised score marginsData Nov 15, 2026Data: Vals AI, Artificial Analysis, benchmarks.bio and 50 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…