Atlas

Compare Models

Gemini 4 Argon vs GPT-5.5

Gemini 4 Argon leads by 5.9 capability points (≈95% interval +3.7 to +8.1): a clear lead over GPT-5.5 · P(higher) ≈ 100%

Wins 133 of 169 shared benchmarks (20 lost, 16 tied)

Capability vs. relative cost
Gemini 4 ArgonGPT-5.5Closed 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-5.5capability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins133 – 20 · 169 shared, 16 tied

Comparing matching setups only (58 benchmarks excluded) · suites counted once
← Gemini 4 ArgonGPT-5.5 →

Gemini 4 ArgonvsGPT-5.5wins by categoryData Nov 15, 2026Data: Vals AI, Center for AI Safety, Artificial Analysis and 36 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is GPT-5.5 minus Gemini 4 Argon.

ARC-AGI-3 (Semi-Private)General QA97.50.4−97.0
BrowseComp Extended (Mercor held-out)Search72.019.0−53.0
Terminal-Bench 4.0 (Artificial Analysis)Code55.15.1−50.0
MirrorCode (ML, +Private, 2L)Code58.910.0−48.9
HieroglyphBenchMultimodal69.722.8−46.9
Time Horizon Index: KSPAgents52.58.3−44.2
ChartographyMultimodal71.631.0−40.6
SRE Bench — Fully SolvedCode44.33.8−40.5
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
ClockBench AccuracyMultimodal83.346.1−37.2
AutomationBench 1.0.6 (Zapier)Agents50.116.6−33.5
Agent Red Teaming k=100 (CAIS)Safety10.241.5−31.3
TxBench Antibody Discovery V1 (100 evaluations)Science63.733.0−30.7
AutomationBench-AA 1.0.6Agents77.547.3−30.2
GameBench 2Games89.259.9−29.2
MASKSafety61.090.1+29.1
TextQuests Game ProgressGames77.348.9−28.4
Furniture Assembly BenchmarkMultimodal72.544.2−28.3
NoCoT-Bench NCRI 15.2General QA151100−28.0
Finance Agent v2 - PrecedentsFinance & Accounting49.821.9−27.9
EnigmaEvalGeneral QA64.637.2−27.4
ALE-BenchCode28052108−24.7
Code MigrationCode68.245.2−23.0
HiL-Bench (Human-in-Loop Benchmark)Agents62.039.7−22.3
GDPval Extended (Mercor held-out rubric score)Professional Work76.454.5−21.9
AA-Briefcase v1.1 EloAgents14851087−21.8
MineBench — Bradley-Terry RatingGames21951846−21.1
Code2Video-AA EloCode12591000−21.1
RiemannBenchMath76.055.2−20.8
SpatialViz-BenchMultimodal94.274.2−20.1
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 - Corporate M&ALegal20.00.0−20.0
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
VariantBench V1 Clean RerunsScience50.330.5−19.8
Finance Agent v2 - Market AnalysisFinance & Accounting79.760.3−19.4
BullshitBench v2 LeaderboardSafety66.047.0−19.0
Terminal-Bench 2.1Code57.776.4+18.7
OncologyBench Strict (25 tasks, October 9 rubric)Health & Medicine33.314.7−18.7
Tax Agent BenchFinance & Accounting79.060.5−18.6
LLM Creative Story WritingChat & Writing0.82.4+18.5
Blueprint-Bench 2Multimodal0.50.4−18.2
LAB-Bench ProtocolQA (SecureBio)Science95.977.9−18.0
HardGeoBench Country AccuracyMultimodal62.545.0−17.5
TxBench Oligo Discovery V1 (113 evaluations)Science51.033.9−17.1
APEX-Agents 1.1Professional Work87.470.3−17.1
SpeciEval (Pooled English Runs)Safety81.398.2+16.9
MindCube TinyMultimodal96.779.9−16.8
Agents' Last Exam (ALE)Agents39.523.0−16.5
Finance Agent v2 - All PassFinance & Accounting55.839.6−16.2
BioSecBench-Function V1 Clean RerunsScience44.428.3−16.1
TextQuests Average Cumulative HarmSafety26.121.4+16.1
ERQAMultimodal86.570.5−16.0
BuseyBench SVGMultimodal7.66.0−15.6
Finance Agent v2 - Disclosure AnalysisFinance & Accounting74.558.9−15.6
LMArena Agent Arena Bash RecoveryAgents-21.49.5+15.5
LiveBench 2.0Agents84.168.7−15.4
APEX-Accounting 1.1Finance & Accounting71.556.3−15.2
HardGeoBench <=100miMultimodal55.040.0−15.0
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine13.928.9+15.0
Finance Agent v2 - AdjustmentsFinance & Accounting60.545.7−14.8
IntPhys 2Multimodal74.659.9−14.7
Humanity's Last Exam DiamondGeneral QA42.527.9−14.6
Legal Research BenchLegal54.840.4−14.4
Artificial Analysis Intelligence Index v4.3.2Indexes52.638.4−14.2
Harvey HLAB — Criteria Pass RateLegal94.080.0−14.1
BioTIER RefusalSafety87.273.4−13.8
DrugDiscoveryBenchScience65.451.6−13.8
HardGeoBench <=10miMultimodal42.528.8−13.8
DeepSWE v1.1Code77.964.4−13.5
GDPval-AA v2.1Professional Work16341335−13.1
SimpleQA Verified (Epoch anti-abstention)General QA76.063.0−13.0
BioSecBench-Surveillance V1 Clean RerunsScience59.847.1−12.7
APEX-SWECode49.236.6−12.6
NoCoT-Bench NCKI (kspine v3)General QA128111−12.2
VCT v2 (SecureBio)Science57.345.3−12.0
Humanity's Last ExamGeneral QA55.643.8−11.8
Finance Agent v2 - Earnings AnalysisFinance & Accounting84.873.2−11.6
Design Arena Overall Code CategoriesCode13261257−11.4
Finance Agent v2 - General Qualitative AnalysisFinance & Accounting77.065.6−11.4
Bio Capabilities Index (SecureBio)Indexes174158−11.4
AA-Omniscience IndexIndexes42.420.5−10.9
Excel Modeling BenchmarkFinance & Accounting75.264.5−10.7
TaxCalcBench TY25 (web search)Finance & Accounting74.064.0−10.0
Machiavelli Deception (CAIS)Safety85.081.7+10.0
HardGeoBench Avg DistanceMultimodal8801196−9.7
MedCodeHealth & Medicine58.849.1−9.7
LMCAGeneral QA63.754.3−9.4
TxBench Preclinical Pharmacology V1 Clean RerunsScience64.755.3−9.3
Public Benefits Bench v1.1Professional Work69.860.9−8.9
ChessBench EloGames14851349−8.1
Mystery Game PuzzlesGames64.056.0−8.0
Finance Agent v2 - General Quantitative AnalysisFinance & Accounting81.573.7−7.8
Artificial Analysis Progress Index v1Indexes342320−7.7
ChessBench CoherenceGames92.6100.0+7.4
HPCT (SecureBio)Science74.266.8−7.4
Finance Agent v2 - ComparablesFinance & Accounting51.043.7−7.3
LMArena Text Arena EloChat & Writing15221481−7.2
MMMU ProMultimodal88.381.2−7.2
LisanBench Path LengthGames75609845+7.1
Conceptual Reasoning IndexIndexes72.467.3−7.0
Finance Agent v2 - Financial ModelingFinance & Accounting29.822.9−6.9
ScBench-Long V1 Clean RerunsScience28.421.6−6.8
ARC-AGI-2General QA91.385.0−6.2
MBCT (SecureBio)Science63.257.1−6.1
VCT (SecureBio)Science58.452.3−6.1
MedXpertQA MM Extended (Mercor held-out)Health & Medicine66.760.8−5.9
GPQA DiamondGeneral QA96.490.7−5.7
LisanBench Difficulty-WeightedGames38493316−5.7
SciCode 1.0.1 (Artificial Analysis)Code61.856.1−5.7
GBENCH One-shot Coding GScoreGames67.962.2−5.6
FrontierMath Tier 4 (v2) - PrivateMath78.072.5−5.5
VCT-Refusal (CAIS)Safety10.716.0+5.3
Epoch Capabilities Index (ECI)Indexes167160−5.2
SpatialBench V1 Clean RerunsScience69.064.1−4.9
AA-LCR v1.1Search79.784.3+4.7
Creative Writing v3 EloChat & Writing17581845+4.4
AA-LCR Extended (Mercor held-out)Search74.770.5−4.2
Chess Puzzles (task version 2.0)Games59.063.0+4.0
ComplexConstraintsChat & Writing53.349.5−3.8
SpatialBench-Long V1 Clean RerunsScience28.632.1+3.6
ARC-AGI-1 Semi-PrivateGeneral QA98.595.0−3.5
ChessBench AccuracyGames81.878.8−3.2
ACCoRDGeneral QA77.380.5+3.2
LLM Chess EloGames16091532−3.2
Vectara Hallucination LeaderboardSafety12.19.3+2.8
LAB-Bench CloningScenarios (SecureBio)Science100.097.3−2.7
Longform Creative WritingChat & Writing80.978.2−2.7
GDP.pdfProfessional Work28.426.0−2.4
SAGEMath53.651.5−2.1
ScBench V1 Clean RerunsScience59.157.1−2.1
ProgramBenchCode2.50.5−2.0
PubMedQA (SecureBio)Health & Medicine82.480.5−1.9
GPQA-Bio (SecureBio)Science89.387.4−1.9
Senior SWE-BenchCode27.729.5+1.8
LegalBenchLegal88.386.5−1.8
CritPt (avg@5)Science27.125.4−1.7
Creative Writing v3 Rubric ScoreChat & Writing15.615.9+1.7
DTBenchGeneral QA97.696.0−1.6
FrontierMath ErdősMath1.50.0−1.5
WCB (SecureBio)Science61.159.7−1.4
GDPval (Mercor 213-task rubric score)Professional Work85.984.6−1.3
PCT Political Inconsistency (CAIS)Safety38.740.0−1.3
AA-AnalystAgent (pass^5)Professional Work48.850.0+1.3
FrontierMath Tiers 1-3 (v2) - PrivateMath86.085.3−0.7
EpiBench V1 Clean RerunsScience28.928.3−0.6
MedScribeHealth & Medicine87.486.9−0.6
DAXBench v2Code88.388.30.0
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 - 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 - 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 - 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
HPCT Refusal (CAIS)Safety29.029.00.0
OTIS Mock AIME 2024-2025Math100.0100.00.0

Gemini 4 ArgonvsGPT-5.5normalised score marginsData Nov 15, 2026Data: Vals AI, Center for AI Safety, Artificial Analysis and 36 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…