Atlas

Compare Models

DeepSeek V4.1 Flash 1017 vs Gemini 4 Argon

Gemini 4 Argon leads by 5.0 capability points (≈95% interval +2.5 to +7.5): a clear lead over DeepSeek V4.1 Flash 1017 · P(higher) ≈ 100%

Wins 135 of 168 shared benchmarks (16 lost, 17 tied)

Capability vs. relative cost
DeepSeek V4.1 Flash 1017Gemini 4 ArgonClosed 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.

DeepSeek V4.1 Flash 1017vsGemini 4 Argoncapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins16 – 135 · 168 shared, 17 tied

Comparing matching setups only (8 benchmarks excluded) · suites counted once
← DeepSeek V4.1 Flash 1017Gemini 4 Argon →

DeepSeek V4.1 Flash 1017vsGemini 4 Argonwins by categoryData Nov 15, 2026Data: Vals AI, Artificial Analysis, SecureBio and 41 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Gemini 4 Argon minus DeepSeek V4.1 Flash 1017.

ARC-AGI-3 (Semi-Private)General QA5.897.5+91.7
HardGeoBench Avg DistanceMultimodal3201741+75.9
AA-PersonalityBench Transparency indexChat & Writing86.721.9−65.4
AA-PersonalityBench Warmth indexChat & Writing27.779.1+51.9
HardGeoBench <=100miMultimodal7.558.8+51.3
HardGeoBench Country AccuracyMultimodal15.063.8+48.7
ArXivMath 10/2026 (Native Harnesses)Math38.383.0+44.7
ArXivMath 09/2026 (Native Harnesses)Math45.387.7+42.5
LisanBench Path LengthGames211077560−42.1
LisanBench Difficulty-WeightedGames77333849−41.6
HardGeoBench <=10miMultimodal3.845.0+41.3
Harvey's Legal Agent Benchmark - Capital MarketsLegal0.040.0+40.0
ArXivMath 08/2026 (Native Harnesses)Math43.982.5+38.6
NoCoT-Bench NCRI 15.2General QA81.6151+38.2
RiemannBench (November 2026 rescoring)Math38.474.4+36.0
Concept Synth INDUCTION Challenge100 Round 1General QA21.057.0+36.0
Terminal-Bench-Science 0.1Science12.948.6+35.7
SimpleQA Verified (Epoch anti-abstention)General QA42.576.0+33.5
Concept Synth INDUCTION Challenge64 Round 1General QA32.865.6+32.8
AA-PersonalityBench Expressiveness indexChat & Writing52.484.8+32.7
AA-PersonalityBench Organization indexChat & Writing88.656.2−32.7
BuseyBench SVGMultimodal4.37.6+32.1
FrontierMath Tier 4 (v2) - PrivateMath46.378.0+31.7
AA-PersonalityBench Conviction indexChat & Writing94.363.8−30.8
AA-PersonalityBench Slop indexChat & Writing39.167.6+28.8
ALE-BenchCode20122805+28.1
Terminal-Bench 4.0Code31.857.6+25.8
Humanity's Sixth SenseMultimodal30.155.4+25.3
Visa Agentic Commerce BenchmarkAgents57.882.2+24.4
Code MigrationCode44.268.2+24.0
Finance Agent v2 - PrecedentsFinance & Accounting26.549.8+23.4
ATLAS (Exa, 200-task subset)Search31.953.8+21.9
Humanity's Last Exam DiamondGeneral QA20.842.5+21.7
Terminal-Bench 4.0 (Artificial Analysis)Code33.855.1+21.2
CritPt v2 (agentic, avg@5)Science52.372.9+20.6
DAYJOB: Data ScienceProfessional Work3.323.3+20.0
Harvey's Legal Agent Benchmark - Antitrust CompetitionLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Corporate GovernanceLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Data Privacy CybersecurityLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Emerging Companies Venture CapitalLegal20.040.0+20.0
Harvey's Legal Agent Benchmark - Funds Asset ManagementLegal20.040.0+20.0
Harvey's Legal Agent Benchmark - International Trade SanctionsLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Litigation Dispute ResolutionLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - Trusts Estates Private ClientLegal20.040.0+20.0
Wiz Cyber Model Arena v3.9Cybersecurity57.677.2+19.6
GameBench 2Games69.989.2+19.3
VIALS-AA (pass^3)Multimodal40.459.6+19.3
AA-Omniscience IndexIndexes4.742.4+18.8
VCT v2 (SecureBio)Science38.557.3+18.8
NoCoT-Bench NCKI (kspine v3)General QA102128+18.6
Build-A-Chip-Bench (20-task index run)Agents0.518.6+18.1
DAYJOB: Finance (October 2026 rescoring)Finance & Accounting12.530.5+18.0
Artificial Analysis Intelligence Index v5.0.1Indexes32.650.4+17.8
Bio Capabilities Index (SecureBio)Indexes149174+17.8
DAYJOB: Healthcare (October 2026 rescoring)Health & Medicine6.023.6+17.6
Finance Agent v2 - Market AnalysisFinance & Accounting62.779.7+17.0
VibeVideoBench (overall quality, no video = 0)Code3.93.1−16.8
τ³-Banking (Mercor run, bm25_grep retrieval)Agents41.658.1+16.5
LMArena Agent Arena Net Improvement (Steering Burden)Agents2.010.7+16.5
FrontierML ($200 wallet)Agents23.539.5+16.1
VCT (SecureBio)Science42.958.4+15.5
Mystery Game PuzzlesGames49.064.0+15.0
FrontierMath Tiers 1-3 (v2) - PrivateMath71.286.0+14.7
GBENCH One-shot Coding GScoreGames53.867.9+14.1
LAB-Bench ProtocolQA (SecureBio)Science81.995.9+14.0
HPCT (SecureBio)Science60.574.2+13.7
Excel Modeling BenchmarkFinance & Accounting61.875.2+13.4
ARC-AGI-2General QA78.191.3+13.2
VIALS-AA (average of 3)Multimodal53.266.3+13.0
Humanity's Last ExamGeneral QA42.655.6+13.0
Chess Puzzles (task version 2.0)Games46.059.0+13.0
HLAB-Vals-1.1Legal14.227.1+12.9
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine26.713.9−12.8
HoneyBench v0.2 (hack rate)Safety76.463.6+12.8
MKA (Artificial Analysis)Professional Work38.551.3+12.7
HANDBOOK.md Agents v1.1Professional Work24.036.6+12.6
APEX-Accounting 1.1Finance & Accounting59.371.5+12.2
Legacy-Bench (Fireworks, avg@3)Code56.344.3−12.0
AA-PersonalityBench Informativeness indexChat & Writing71.482.9+11.5
WCB (SecureBio)Science49.661.1+11.5
LMArena Agent Arena Praise vs ComplaintAgents1.224.1+11.5
τ³-Banking (Fireworks, banking_knowledge)Agents46.858.2+11.4
DAXBench v2Code77.788.3+10.6
Finance Agent v2 - All PassFinance & Accounting45.255.8+10.5
DAYJOB: AccountingFinance & Accounting7.618.0+10.4
BrokenArXiv 10/2026 (Native Harnesses)Math49.739.8−9.9
Legal Research BenchLegal45.254.8+9.6
AA-PersonalityBench Imagination indexChat & Writing37.246.7+9.6
Finance Agent v2 - Earnings AnalysisFinance & Accounting75.384.8+9.5
MBCT (SecureBio)Science53.963.2+9.3
AutomationBench-AA 1.0.6Agents68.577.5+9.0
DualEntry Accounting AI BenchmarkFinance & Accounting82.291.1+8.9
Artificial Analysis Coding Agent Index v2.1Indexes36.845.5+8.7
MMMU ProMultimodal79.788.3+8.6
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Legal18.827.4+8.6
GDP.pdf (Artificial Analysis, All-pass)Professional Work17.225.6+8.4
Finance Agent v2 - Disclosure AnalysisFinance & Accounting66.374.5+8.2
FrontierCode 2.0 - Pass RateCode22.430.6+8.2
APEX-Agents 1.1Professional Work79.487.4+8.0
FrontierCode 2.0 - ScoreCode19.827.6+7.8
o11y-bench 1.1 Pass^3Agents79.787.5+7.8
Artificial Analysis Intelligence Index v4.3.2Indexes45.152.6+7.5
BrokenArXiv 08/2026 (Native Harnesses)Math46.754.2+7.4
CWE-bench v2 Pass@1Cybersecurity57.665.0+7.4
Finance Agent v2 - Financial ModelingFinance & Accounting22.829.8+7.1
Bedside BenchHealth & Medicine85.792.7+7.0
Artificial Analysis Progress Index v1Indexes324342+6.3
VIALS-AA (pass@3)Multimodal66.572.7+6.2
Design Arena Overall Code CategoriesCode12891326+6.1
WARC-Bench (192-task test split)Agents70.576.6+6.1
SciCode 1.0.1 (Artificial Analysis)Code55.961.8+5.9
Epoch Capabilities Index (ECI)Indexes159167+5.8
Excel Modeling Benchmark v1.1Finance & Accounting69.274.8+5.7
LMArena Agent Arena Confirmed SuccessAgents2.313.2+5.5
CursorBench 4.0Code47.752.9+5.2
LAB-Bench CloningScenarios (SecureBio)Science94.8100.0+5.2
Vals Finance IndexIndexes60.966.0+5.1
Vals Legal IndexIndexes71.976.8+4.9
LiveBench 2.0Agents79.284.1+4.9
Finance Agent v2.1Finance & Accounting62.967.7+4.7
FrontierPhysics v0.1Science3.68.3+4.7
Access to Justice BenchmarkLegal73.168.5−4.6
LMArena Agent Arena Steering BurdenAgents4.013.2+4.6
Harvey HLAB — Criteria Pass RateLegal89.794.0+4.3
Tax Agent BenchFinance & Accounting74.979.0+4.1
Agents' Last Exam v1.1 (ALE v1.1)Agents36.140.2+4.1
ARC-AGI-1 Semi-PrivateGeneral QA94.598.5+4.0
LMArena Agent Arena Bash ReliabilityAgents2.4-5.4−3.9
AA-Briefcase v1.1 EloAgents14161485+3.8
DAYJOB: LawLegal6.29.9+3.7
WMDP-Bio (SecureBio)Science87.991.5+3.6
Finance Agent v2 - ComparablesFinance & Accounting47.551.0+3.4
Code Migration v1.1Code67.570.8+3.3
CritPt (avg@5)Science24.027.1+3.1
AA-LCR v1.1Search82.779.7−3.0
Finance Agent v2 - General Qualitative AnalysisFinance & Accounting74.177.0+2.9
LMArena Code Arena WebDevCode16641682+2.4
PubMedQA (SecureBio)Health & Medicine80.082.4+2.4
STEMBench120 v1.0.0 (closed-book)Science2.52.4−2.3
Finance Agent v2 - General Quantitative AnalysisFinance & Accounting79.281.5+2.3
Finance Agent v2 - AdjustmentsFinance & Accounting58.460.5+2.2
DAYJOB: TaxFinance & Accounting1.23.2+2.0
Bug Hunt BenchCode40.738.7−2.0
GDPxProfessional Work5.47.3+1.9
GPQA-Bio (SecureBio)Science87.889.3+1.5
MMLU-Bio (SecureBio)Science90.792.1+1.4
DeepSWE v1.1Code77.077.9+0.9
OTIS Mock AIME 2024-2025Math99.4100.0+0.6
FrontierCode 1.1 Main (Artificial Analysis)Code43.743.5−0.2
LMArena Agent Arena Tool HallucinationAgents0.30.5+0.1
APEX-SWECode49.149.2+0.1
BrokenArXiv 09/2026 (Native Harnesses)Math49.749.70.0
ErdosBenchMath3.03.00.0
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 - Healthcare Life SciencesLegal20.020.00.0
Harvey's Legal Agent Benchmark - ImmigrationLegal20.020.00.0
Harvey's Legal Agent Benchmark - InsuranceLegal0.00.00.0
Harvey's Legal Agent Benchmark - Intellectual PropertyLegal20.020.00.0
Harvey's Legal Agent Benchmark - 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

DeepSeek V4.1 Flash 1017vsGemini 4 Argonnormalised score marginsData Nov 15, 2026Data: Vals AI, Artificial Analysis, SecureBio and 41 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…