Atlas

Compare Models

Gemini 4 Argon vs GPT-6.1 Sol

GPT-6.1 Sol leads by 0.4 capability points (≈95% interval −2.2 to +2.9): statistically tied · P(higher) ≈ 61%

Gemini 4 Argon wins 143 of 275 shared benchmarks (117 lost, 15 tied)

Capability vs. relative cost
Gemini 4 ArgonGPT-6.1 SolClosed 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 ArgonvsGPT-6.1 Solcapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins143 – 117 · 275 shared, 15 tied

Comparing matching setups only (84 benchmarks excluded) · suites counted once
← Gemini 4 ArgonGPT-6.1 Sol →

Gemini 4 ArgonvsGPT-6.1 Solwins by categoryData Nov 15, 2026Data: Vals AI, Artificial Analysis, benchmarks.bio and 55 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is GPT-6.1 Sol minus Gemini 4 Argon.

AA-PersonalityBench Transparency indexChat & Writing21.992.4+71.2
SAFE-Teen 106 worst branch: Health Advice (no prompt)Safety70.88.8+62.0
AA-PersonalityBench Informativeness indexChat & Writing82.927.7−55.8
AA-PersonalityBench Slop indexChat & Writing67.614.3−53.8
SAFE-Teen 106: Health Advice (no prompt)Safety57.44.7+52.8
SAFE-Teen Health Advice (no prompt)Safety56.24.2+52.0
SAFE-Teen 106 worst branch: Relationships (no prompt)Safety65.719.1+46.5
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 - Funds Asset ManagementLegal40.00.0−40.0
Harvey's Legal Agent Benchmark - Trusts Estates Private ClientLegal40.00.0−40.0
CyberBench v1.1 OverallCybersecurity77.939.3−38.6
HieroglyphBenchMultimodal69.731.4−38.3
MASKSafety61.098.3+37.3
SAFE-Teen 106 worst branch: Urgent Safety (no prompt)Safety52.615.4+37.3
SAFE-Teen 106 worst branch: Health Advice (under-18 prompt)Safety43.67.7+35.9
AA-PersonalityBench Expressiveness indexChat & Writing84.850.5−34.6
Concept Synth INDUCTION Challenge100 Round 1General QA57.090.0+33.0
Agent Initiative BenchmarkAgents5.68.7+30.5
SAFE-Teen Urgent Safety (no prompt)Safety33.33.5+29.8
SAFE-Teen 106: Urgent Safety (no prompt)Safety33.13.4+29.7
Concept Synth INDUCTION Challenge64 Round 1General QA65.695.3+29.7
Terminal-Bench 2.1Code57.785.4+27.7
SAFE-Teen Health Advice (under-18 prompt)Safety29.62.0+27.6
SAFE-Teen 106: Health Advice (under-18 prompt)Safety30.83.3+27.4
SnorkelLegalLegal46.020.7−25.3
AA-PersonalityBench Organization indexChat & Writing56.281.0+25.0
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine13.938.3+24.4
Finance Agent v2 - PrecedentsFinance & Accounting49.825.8−24.0
FrontierMath Tier 4 (v2) - PrivateMath78.0100.0+22.0
DAYJOB: LawLegal9.931.8+21.9
HLAB-Vals-1.1Legal27.16.3−20.8
Harvey's Legal Agent Benchmark - Antitrust CompetitionLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - Corporate GovernanceLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - Data Privacy CybersecurityLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - Healthcare Life SciencesLegal20.00.0−20.0
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 - Structured Finance SecuritizationLegal0.020.0+20.0
Visa Agentic Commerce BenchmarkAgents82.262.2−20.0
Machiavelli Deception (CAIS)Safety85.078.8+18.7
SAFE-Teen Relationships (no prompt)Safety28.610.0+18.7
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Legal27.48.8−18.6
SAFE-Teen 106: Relationships (no prompt)Safety28.510.2+18.3
sudo L7 (October 2026 rescoring)Code16.133.9+17.8
SAFE-Teen 106 worst branch: Urgent Safety (under-18 prompt)Safety30.212.7+17.5
OpenWorldEngineer (Mean@8)Code24.141.2+17.1
Tax Agent BenchFinance & Accounting79.062.3−16.7
Legal Research BenchLegal54.838.5−16.3
GDP.xlsxProfessional Work38.322.0−16.3
Humanity's Last Exam DiamondGeneral QA42.558.8+16.3
RareDiseaseBench V1 (35 evaluations)Science43.827.6−16.2
SpeciEval (Pooled English Runs)Safety81.397.4+16.1
Finance Agent v2 - Disclosure AnalysisFinance & Accounting74.558.4−16.1
Mystery Game PuzzlesGames64.080.0+16.0
WellnessBenchSafety50.066.0+16.0
Finance Agent v2 - All PassFinance & Accounting55.839.9−15.9
Finance Agent v2 - AdjustmentsFinance & Accounting60.544.7−15.8
ATLAS (Exa, 200-task subset)Search53.869.5+15.7
AA-PersonalityBench Imagination indexChat & Writing46.761.9+15.4
HardGeoBench Avg DistanceMultimodal7411239−15.4
Finance Agent v2 - Market AnalysisFinance & Accounting79.764.5−15.2
τ³-Banking (Mercor run, bm25_grep retrieval)Agents58.143.0−15.1
HardGeoBench <=100miMultimodal58.843.8−15.0
Ramp Accounting BenchFinance & Accounting61.146.4−14.8
APEX-Agents 1.1Professional Work87.473.2−14.2
TxBench Antibody Discovery V1 (100 evaluations)Science63.749.7−14.0
CheatBenchSafety28.842.8−14.0
Finance Agent v2 - General Qualitative AnalysisFinance & Accounting77.063.2−13.8
τ³-Banking (Fireworks, banking_knowledge)Agents58.244.7−13.6
AA-PersonalityBench Warmth indexChat & Writing79.165.7−13.5
FrontierMemory v1.1Agents64.451.1−13.3
Finance Agent v2 - Earnings AnalysisFinance & Accounting84.871.5−13.3
HPCT Refusal (CAIS)Safety29.042.0+13.0
LMArena Agent Arena Bash RecoveryAgents-21.44.5+13.0
SWE-bench Multimodal (Vals v1, 480 Tasks)Code56.043.5−12.5
APEX-1 (Mercor, mean score)Professional Work81.169.3−11.8
Terminal-Bench-Science 0.1Science48.637.1−11.4
SpatialBench-Long V1 Clean RerunsScience28.639.9+11.3
Dual-Arm Sim (task completion)Agents11.30.0−11.2
HardGeoBench Country AccuracyMultimodal63.852.5−11.2
GBENCH One-shot Coding GScoreGames67.979.0+11.1
BioSecBench-Refusal V1 Clean RerunsSafety48.737.6−11.1
FrontierCode 2.0 (Artificial Analysis)Code25.736.7+11.0
AutomationBench-AA 1.0.6Agents77.566.6−10.9
TextQuests Average Cumulative HarmSafety26.129.3−10.9
Legacy-Bench (Fireworks, avg@3)Code44.355.0+10.7
EBR-bench v4 (card ban, single agent)Games64.854.3−10.5
Public Benefits Bench v1.1Professional Work69.859.3−10.4
SAFE-Teen Urgent Safety (under-18 prompt)Safety12.32.1+10.3
Vals Finance IndexIndexes66.055.8−10.2
SAFE-Teen 106: Urgent Safety (under-18 prompt)Safety12.12.0+10.1
HardGeoBench <=10miMultimodal45.035.0−10.0
MedCodeHealth & Medicine58.848.8−10.0
DualEntry Accounting AI BenchmarkFinance & Accounting91.181.2−9.9
WARC-Bench (192-task test split)Agents76.686.3+9.7
AA-PersonalityBench Conviction indexChat & Writing63.854.3−9.6
SAFE-Teen 106 worst branch: Relationships (under-18 prompt)Safety14.04.5+9.5
HealthBench (length-adjusted, Artificial Analysis)Health & Medicine49.158.5+9.5
EQ-Bench 4Chat & Writing12361272+9.5
OncologyBench Strict (25 tasks, October 9 rubric)Health & Medicine33.324.0−9.3
CritPt v2 (agentic, avg@5)Science72.982.0+9.1
BioSecBench-Surveillance V1 Clean RerunsScience59.850.8−9.0
Chess Puzzles (task version 2.0)Games59.068.0+9.0
Code2Video-AA EloCode12591369+8.9
ClockBench AccuracyMultimodal83.374.4−8.9
Humanity's Sixth SenseMultimodal55.446.6−8.8
Terminal-Bench Science 0.1 (Vals)Science44.352.9+8.6
Agent Red Teaming k=100 (CAIS)Safety10.218.8−8.6
Arena Alignment IndexSafety79.487.9+8.5
SnorkelFinance 2.0Finance & Accounting15.16.6−8.5
Finance Agent v2 - ComparablesFinance & Accounting51.042.6−8.3
ERQAMultimodal86.578.2−8.3
Vals Legal IndexIndexes76.868.7−8.1
HANDBOOK.md Agents v1.1Professional Work36.628.6−8.0
FrontierMath Tiers 1-3 (v2) - PrivateMath86.093.7+7.7
MindCube TinyMultimodal96.789.1−7.6
Furniture Assembly BenchmarkMultimodal72.580.0+7.5
APEX-Accounting 1.1Finance & Accounting71.564.3−7.2
DAYJOB: AccountingFinance & Accounting18.025.2+7.2
Snorkel Digital Work IndexIndexes28.321.2−7.1
SAGEMath53.646.5−7.1
Terminal-Bench 4.0 (Artificial Analysis)Code55.148.0−7.1
LMArena Text Arena EloChat & Writing15221482−7.0
Toloka Arena (v3)Agents73.066.0−7.0
Finance Agent v2 - General Quantitative AnalysisFinance & Accounting81.574.5−7.0
Finance Agent v2 - Financial ModelingFinance & Accounting29.823.1−6.7
GDP.pdfProfessional Work28.435.0+6.6
ALE-BenchCode28052990+6.6
Code Migration v1.1Code70.877.3+6.5
Snorkel SWE-bench CLICode16.39.8−6.5
SRE Bench — Fully SolvedCode44.350.8+6.5
BioSecBench-Structure V1 (58 evaluations)Science14.921.3+6.3
SnorkelRevOpsProfessional Work18.412.1−6.3
Harvey HLAB — Criteria Pass RateLegal94.087.8−6.3
Snorkel WorkplaceAgentsProfessional Work21.215.0−6.2
SciCode 1.0.1 (Artificial Analysis)Code61.855.8−6.0
TextQuests Game ProgressGames77.371.4−5.9
Finance Agent v2.1Finance & Accounting67.761.8−5.8
Westworld Finance Diligence BenchFinance & Accounting60.755.0−5.8
DistressBenchSafety68.674.2+5.5
Decision-Bench v1.1Professional Work54.549.1−5.5
VCT-Refusal (CAIS)Safety10.716.1+5.4
LMArena Agent Arena Bash ReliabilityAgents-5.45.0+5.2
BioSecBench-RespiratoryVirus-Surveillance V1 (110 evaluations)Science17.612.4−5.2
ChessBench CoherenceGames92.687.5−5.1
EnigmaEvalGeneral QA64.659.5−5.1
CrossSearch (pass^3)Search64.059.0−5.0
TxBench Preclinical Pharmacology V1 Clean RerunsScience64.759.7−5.0
Design Arena Overall Code CategoriesCode13261356+5.0
GDPval-AA v2.1Professional Work16341521−5.0
RiemannBench (November 2026 rescoring)Math74.479.2+4.8
DAYJOB: Data ScienceProfessional Work23.318.7−4.6
Financial Coaching BenchFinance & Accounting73.168.5−4.6
ChessBench EloGames14851562+4.6
CritPt (avg@5)Science27.131.7+4.6
Excel Modeling BenchmarkFinance & Accounting75.270.8−4.5
ScBench V1 Clean RerunsScience59.163.6+4.4
AA-LCR v1.1Search79.784.0+4.3
Humanity's Last ExamGeneral QA55.651.3−4.3
BrowseComp Extended (Mercor held-out)Search72.067.7−4.3
Vals Science IndexIndexes55.459.6+4.3
DAXBench v2Code88.392.6+4.3
VariantBench V1 Clean RerunsScience50.346.0−4.2
SWE Atlas v1.1 - OverallCode70.165.9−4.2
Artificial Analysis Intelligence Index v5.0.1Indexes50.454.6+4.2
HoneyBench v0.2 (hack rate)Safety63.659.5+4.1
HealthBench ProfessionalHealth & Medicine64.268.3+4.1
RCM-Bench-1Health & Medicine91.587.4−4.1
TaxCalcBench TY25 (web search)Finance & Accounting74.070.0−4.0
SpatialViz-BenchMultimodal94.290.3−4.0
Web Clone BenchCode8.612.6+4.0
BuseyBench SVGMultimodal7.68.0+3.9
MacroscopeBenchCode72.876.7+3.9
Excel Modeling Benchmark v1.1Finance & Accounting74.871.1−3.7
AA-3D v1.0 EloCode11471121−3.7
Professional Reasoning Benchmark - FinanceFinance & Accounting49.946.2−3.7
LisanBench Path LengthGames75608736+3.7
ACCoRDGeneral QA77.380.9+3.6
MKA (Artificial Analysis)Professional Work51.347.7−3.5
MetagenomicsBench V1Science42.345.7+3.3
BioMysteryBench (Vals Terminus 2)Science76.379.6+3.3
TxBench-Antibody-Development V1 (60 evaluations)Science31.127.8−3.3
DataCenterBench Tier 1Agents35.038.3+3.3
LMArena Agent Arena Net Improvement (Steering Burden)Agents10.712.5+3.3
MultiChallengeChat & Writing78.782.0+3.3
MMMU ProMultimodal88.385.1−3.2
DAYJOB: HealthcareHealth & Medicine10.87.6−3.2
FrontierCode 1.1 Main (Artificial Analysis)Code43.546.6+3.2
o11y-bench 1.1 Pass^3Agents87.584.4−3.1
IOI (Vals v2)Code100.096.9−3.1
SWE-Bench Pro V2 — FullCode99.296.1−3.1
HealthBenchHealth & Medicine60.056.9−3.1
BioSimBench 1.0Science18.415.3−3.1
MissionReportProfessional Work64.161.0−3.1
Code MigrationCode68.265.1−3.1
FrontierMath ErdősMath1.54.4+2.9
ARC-AGI-2General QA91.394.2+2.9
NeuroscienceBench V1 (91 evaluations)Science19.122.0+2.9
GameBench 2Games89.292.0+2.8
BigFinanceBench private holdout (final-answer accuracy)Finance & Accounting54.752.0−2.7
Access to Justice BenchmarkLegal68.565.8−2.6
Artificial Analysis Progress Index v1Indexes342348+2.3
ARC-AGI-3 (Semi-Private)General QA97.395.0−2.3
IntPhys 2Multimodal74.672.3−2.3
APEX-SWECode49.246.9−2.3
SAFE-Teen 106: Relationships (under-18 prompt)Safety3.00.9+2.1
SimpleQA Verified (Epoch anti-abstention)General QA76.073.9−2.1
SAFE-Teen Relationships (under-18 prompt)Safety3.00.9+2.1
MedXpertQA MM Extended (Mercor held-out)Health & Medicine66.764.7−2.0
VIALSMultimodal47.249.1+1.9
Epoch Capabilities Index (ECI)Indexes167169+1.9
SpatialBench V1 Clean RerunsScience69.067.3−1.7
ScBench-Long V1 Clean RerunsScience28.426.7−1.7
SnorkelUnderwrite 2.0Finance & Accounting23.325.0+1.7
LMArena Agent Arena Praise vs ComplaintAgents24.127.5+1.7
LisanBench Difficulty-WeightedGames38494004+1.7
BioSecBench-Function V1 Clean RerunsScience44.446.0+1.6
Professional Reasoning Benchmark - LegalLegal49.447.9−1.5
TxBench Oligo Discovery V1 (113 evaluations)Science51.052.5+1.5
DeepSWE v1.1 (Fireworks, avg@3)Code67.869.3+1.5
PCT Political Inconsistency (CAIS)Safety38.737.3+1.4
ChessBenchGames70.068.7−1.3
LiveBench 2.0Agents84.185.4+1.3
AA-AnalystAgent (pass^5)Professional Work48.850.0+1.3
Political Science LLM BenchmarkProfessional Work69.668.3−1.2
Bedside BenchHealth & Medicine92.793.9+1.2
ChessBench AccuracyGames81.880.7−1.2
NoCoT-Bench NCRI 15.2General QA151149−1.2
GDPxProfessional Work7.38.4+1.1
DTBenchGeneral QA97.696.5−1.1
SAGE v1.1Math89.590.5+1.0
BullshitBench v2 LeaderboardSafety66.065.0−1.0
Snorkel Agentic Coding 2.0Code41.842.8+1.0
MedScribeHealth & Medicine87.486.5−1.0
LMCAGeneral QA63.762.8−0.9
MysteryMechanismScience45.546.4+0.9
AA-Briefcase v1.1 EloAgents14851469−0.9
NoCoT-Bench NCKI (kspine v3)General QA128130+0.8
DAYJOB: FinanceFinance & Accounting18.818.0−0.8
DAYJOB: TaxFinance & Accounting3.24.0+0.8
LMArena Agent Arena Confirmed SuccessAgents13.211.7−0.8
Artificial Analysis Intelligence Index v4.3.2Indexes52.651.8−0.7
TissueMicroscopyBench V1 (49 evaluations)Science21.822.4+0.7
EpiBench V1 Clean RerunsScience28.928.3−0.6
Terminal-Bench 4.0Code57.658.2+0.6
ProgramBenchCode2.53.0+0.5
Prophet Arena Arena Score (Fixed Context)Forecasting85.384.8−0.5
AA-Omniscience IndexIndexes42.441.5−0.4
Conceptual Reasoning IndexIndexes72.472.7+0.4
HiL-Bench (Human-in-Loop Benchmark)Agents62.062.3+0.3
Immigration BenchLegal68.167.8−0.3
LMArena Agent Arena Tool HallucinationAgents0.51.0+0.2
ComplexConstraints (November 2026 rescoring)Chat & Writing70.370.1−0.2
LMArena Agent Arena Steering BurdenAgents13.213.4+0.1
SnorkelManufacturingProfessional Work19.519.6+0.1
Wiz Cyber Model Arena v3.9Cybersecurity70.870.7−0.1
LLM Chess EloGames16091612+0.1
Prophet Arena Arena Score (Agentic, OpenCode)Forecasting86.086.00.0
ARC-AGI-1 Semi-PrivateGeneral QA98.598.50.0
Harvey's Legal Agent Benchmark - Arbitration International Dispute ResolutionLegal0.00.00.0
Harvey's Legal Agent Benchmark - Banking FinanceLegal0.00.00.0
Harvey's Legal Agent Benchmark - Bankruptcy RestructuringLegal0.00.00.0
Harvey's Legal Agent Benchmark - Corporate M&ALegal20.020.00.0
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 - InsuranceLegal0.00.00.0
Harvey's Legal Agent Benchmark - Litigation Dispute ResolutionLegal0.00.00.0
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
OTIS Mock AIME 2024-2025Math100.0100.00.0
ProofBench v1.1Math99.099.00.0

Gemini 4 ArgonvsGPT-6.1 Solnormalised score marginsData Nov 15, 2026Data: Vals AI, Artificial Analysis, benchmarks.bio 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…