Atlas

Compare Models

DeepSeek V4.1 Flash 1017 vs Grok 4.6

DeepSeek V4.1 Flash 1017 leads by 0.9 capability points (≈95% interval −1.3 to +3.2): statistically tied · P(higher) ≈ 79%

Wins 65 of 125 shared benchmarks (46 lost, 14 tied)

Capability vs. relative cost
DeepSeek V4.1 Flash 1017Grok 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.

DeepSeek V4.1 Flash 1017vsGrok 4.6capability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins65 – 46 · 125 shared, 14 tied

Comparing matching setups only (6 benchmarks excluded) · suites counted once
← DeepSeek V4.1 Flash 1017Grok 4.6 →

DeepSeek V4.1 Flash 1017vsGrok 4.6wins by categoryData Nov 15, 2026Data: Vals AI, Artificial Analysis, SecureBio and 27 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 DeepSeek V4.1 Flash 1017.

AA-PersonalityBench Organization indexChat & Writing88.614.3−75.0
HardGeoBench Avg DistanceMultimodal32011807+43.0
HardGeoBench Country AccuracyMultimodal15.043.8+28.8
AA-PersonalityBench Informativeness indexChat & Writing71.444.8−26.9
HardGeoBench <=100miMultimodal7.532.5+25.0
LAB-Bench LitQA2 (SecureBio)Science49.028.9−20.1
Harvey's Legal Agent Benchmark - Antitrust CompetitionLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Banking FinanceLegal0.020.0+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 CapitalLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - Energy Natural ResourcesLegal20.040.0+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 - Structured Finance SecuritizationLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Trusts Estates Private ClientLegal20.00.0−20.0
AA-PersonalityBench Imagination indexChat & Writing37.220.0−17.3
HardGeoBench <=10miMultimodal3.820.0+16.3
AA-PersonalityBench Transparency indexChat & Writing86.771.4−15.4
Mystery Game PuzzlesGames49.034.0−15.0
Wiz Cyber Model Arena v3.9Cybersecurity60.345.5−14.8
FrontierMath Tier 4 (v2) - PrivateMath46.331.7−14.6
DeepsecBench — Revalidated V1 (233 issues)Code30.016.1−13.9
LAB-Bench ProtocolQA (SecureBio)Science81.968.1−13.8
Bug Hunt BenchCode40.727.3−13.4
Chess Puzzles (task version 2.0)Games46.033.0−13.0
AA-Omniscience IndexIndexes4.730.5+12.9
AA-PersonalityBench Expressiveness indexChat & Writing52.463.8+11.5
Vibe Code Bench v1.1Code87.776.2−11.5
ARC-AGI-2General QA78.167.1−11.0
CritPt v2 (agentic, avg@5)Science52.362.3+10.0
DeepSWE v1.1Code77.067.5−9.5
DAYJOB: Healthcare (October 2026 rescoring)Health & Medicine6.014.8+8.8
Bio Capabilities Index (SecureBio)Indexes149137−8.7
FrontierCode 2.0 (Artificial Analysis)Code28.519.8−8.7
APEX-SWECode49.157.7+8.6
DAXBench v2Code77.786.2+8.5
DAYJOB: Finance (October 2026 rescoring)Finance & Accounting12.521.0+8.5
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine26.718.3−8.3
FoodTruck BenchGames5328959537+8.0
Terminal-Bench 4.0 (Artificial Analysis)Code28.821.2−7.6
EnterpriseBench: CoreCraftProfessional Work58.565.6+7.1
NoCoT-Bench NCKI (kspine v3)General QA102112+7.0
ARC-AGI-1 Semi-PrivateGeneral QA94.587.5−7.0
SimpleQA Verified (Epoch anti-abstention)General QA42.549.3+6.8
ALE-BenchCode20121822−6.7
CursorBench 4.0Code47.741.4−6.3
AA-Briefcase v1.1 EloAgents14161526+6.0
ZeroBench (Main questions, pass@5)Multimodal23.017.0−6.0
LMArena Code Arena WebDevCode16641618−6.0
Terminal-Bench 4.0Code31.025.3−5.7
TCSExamMath46.040.5−5.5
VCT (SecureBio)Science42.948.3+5.4
FrontierMath Tiers 1-3 (v2) - PrivateMath71.266.0−5.3
Vals Index v2.1Indexes56.652.1−4.5
CritPt (avg@5)Science24.019.7−4.3
Terminal-Bench-Science 0.1Science10.56.2−4.3
LiveBench 2.0Agents79.274.9−4.3
LMArena Agent Arena Net Improvement (Steering Burden)Agents2.0-0.2−4.1
τ³-Banking (Mercor run, bm25_grep retrieval)Agents41.645.7+4.1
Concept Synth INDUCTION Challenge100 Round 1General QA21.025.0+4.0
τ³-Banking (Fireworks, banking_knowledge)Agents46.850.7+3.9
AA-PersonalityBench Slop indexChat & Writing39.135.3−3.8
AA-PersonalityBench Warmth indexChat & Writing27.731.5+3.8
LMArena Agent Arena Bash ReliabilityAgents2.410.0+3.8
ARC-AGI-3 (Semi-Private)General QA5.82.1−3.7
RedlineBenchLegal44.948.5+3.6
Tax Agent BenchFinance & Accounting74.971.4−3.5
MBCT (SecureBio)Science53.950.4−3.5
Finance Agent v2Finance & Accounting57.053.7−3.3
Concept Synth INDUCTION Challenge64 Round 1General QA32.835.9+3.1
LMArena Agent Arena Steering BurdenAgents4.0-1.8−2.9
Legal Research BenchLegal45.248.1+2.9
Humanity's Last ExamGeneral QA42.639.7−2.9
NoCoT-Bench NCRI 15.2General QA81.686.8+2.8
Harvey HLAB — Criteria Pass RateLegal89.792.5+2.8
BuseyBench SVGMultimodal4.34.6+2.6
LMArena Agent Arena Confirmed SuccessAgents2.3-2.6−2.5
Bedside BenchHealth & Medicine85.783.3−2.4
APEX-Agents 1.1Professional Work79.477.3−2.1
AA-PersonalityBench Conviction indexChat & Writing94.392.4−1.9
LMArena Agent Arena Praise vs ComplaintAgents1.2-2.5−1.8
Access to Justice BenchmarkLegal73.171.3−1.8
AA-LCR v1.1Search82.781.0−1.7
GBENCH One-shot Coding GScoreGames53.855.4+1.6
Artificial Analysis Intelligence Index v5.0.1Indexes32.631.0−1.6
AutomationBench-AA 1.0.6Agents68.567.0−1.5
GameBench 2Games69.968.7−1.2
Epoch Capabilities Index (ECI)Indexes159158−1.2
HPCT (SecureBio)Science60.561.6+1.1
MMLU-Bio (SecureBio)Science90.791.8+1.1
APEX-Accounting 1.1Finance & Accounting59.360.3+1.0
Excel Modeling BenchmarkFinance & Accounting61.862.7+0.9
ZeroBench Main Mean pass@1 (five samples)Multimodal11.811.0−0.8
MMMU ProMultimodal79.780.5+0.8
Artificial Analysis Intelligence Index v4.3.2Indexes45.144.3−0.8
Artificial Analysis Progress Index v1Indexes324322−0.8
Humanity's Last Exam DiamondGeneral QA20.821.5+0.7
WCB (SecureBio)Science49.650.3+0.7
GDP.pdf (Artificial Analysis, All-pass)Professional Work17.217.8+0.6
APEX-1 (Mercor, mean score)Professional Work68.367.7−0.6
LAB-Bench CloningScenarios (SecureBio)Science94.894.2−0.6
SciCode 1.0.1 (Artificial Analysis)Code55.956.5+0.6
ReactBench (39 tasks)Code34.433.8−0.5
Arena Image-to-WebDevCode16091607−0.5
GPQA-Bio (SecureBio)Science87.887.4−0.4
LMArena Agent Arena Tool HallucinationAgents0.31.0+0.4
Code MigrationCode44.244.6+0.4
OTIS Mock AIME 2024-2025Math99.499.2−0.3
Design Arena Overall Code CategoriesCode12891288−0.2
HealthBench Professional (Length-Adjusted)Health & Medicine51.751.7−0.1
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 - Capital MarketsLegal0.00.00.0
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 - Environmental ESGLegal0.00.00.0
Harvey's Legal Agent Benchmark - Funds Asset ManagementLegal20.020.00.0
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 - International Trade SanctionsLegal0.00.00.0
Harvey's Legal Agent Benchmark - Litigation Dispute ResolutionLegal20.020.00.0
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

DeepSeek V4.1 Flash 1017vsGrok 4.6normalised score marginsData Nov 15, 2026Data: Vals AI, Artificial Analysis, SecureBio and 27 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…