Atlas

Compare Models

DeepSeek V4.1 Flash 1017 vs GPT-5.6 Terra

DeepSeek V4.1 Flash 1017 leads by 0.3 capability points (≈95% interval −2.0 to +2.5): statistically tied · P(higher) ≈ 59%

Wins 59 of 128 shared benchmarks (54 lost, 15 tied)

Capability vs. relative cost
DeepSeek V4.1 Flash 1017GPT-5.6 TerraClosed 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 1017vsGPT-5.6 Terracapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins59 – 54 · 128 shared, 15 tied

Comparing matching setups only (18 benchmarks excluded) · suites counted once
← DeepSeek V4.1 Flash 1017GPT-5.6 Terra →

DeepSeek V4.1 Flash 1017vsGPT-5.6 Terrawins by categoryData Nov 15, 2026Data: Vals AI, Artificial Analysis, SecureBio and 32 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is GPT-5.6 Terra minus DeepSeek V4.1 Flash 1017.

LisanBench Difficulty-WeightedGames77331197−70.0
LisanBench Path LengthGames211073303−55.3
HardGeoBench Avg DistanceMultimodal32011991+37.3
HardGeoBench Country AccuracyMultimodal15.045.0+30.0
HardGeoBench <=100miMultimodal7.536.3+28.8
FrontierMath Tier 4 (v2) - PrivateMath46.370.7+24.4
HardGeoBench <=10miMultimodal3.825.0+21.3
Harvey's Legal Agent Benchmark - Corporate M&ALegal20.00.0−20.0
Harvey's Legal Agent Benchmark - Emerging Companies Venture CapitalLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - Energy Natural ResourcesLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - Funds Asset ManagementLegal20.00.0−20.0
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 - Litigation Dispute ResolutionLegal20.00.0−20.0
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
CritPt v2 (agentic, avg@5)Science50.669.7+19.1
APEX-SWECode49.131.1−18.0
BuseyBench SVGMultimodal4.36.1+17.3
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Legal18.82.3−16.5
GameBench 2Games69.953.5−16.4
FrontierMath Tiers 1-3 (v2) - PrivateMath71.286.0+14.7
Mystery Game PuzzlesGames49.035.0−14.0
Vibe Code Bench v1.1Code87.774.6−13.1
DeepSWE v1.1Code77.064.8−12.2
AA-Briefcase v1.1 EloAgents14161202−11.7
DeepsecBench — Revalidated V1 (233 issues)Code30.019.1−10.9
VoxelBench Text RatingGames20471894−10.9
WCB (SecureBio)Science49.659.8+10.2
Harvey HLAB — Criteria Pass RateLegal89.779.8−10.0
Bug Hunt BenchCode40.230.5−9.7
Tax Agent BenchFinance & Accounting74.965.2−9.7
GBENCH One-shot Coding GScoreGames53.863.3+9.5
DAYJOB: Finance (October 2026 rescoring)Finance & Accounting12.53.3−9.2
Terminal-Bench 4.0Code31.822.7−9.1
CharXiv ReasoningMultimodal77.085.9+8.9
AutomationBench-AA 1.0.6Agents68.559.6−8.8
Finance Agent v2 - AdjustmentsFinance & Accounting58.449.7−8.7
Design Arena Overall Code CategoriesCode12891238−8.5
Humanity's Last Exam DiamondGeneral QA20.828.8+8.0
HPCT (SecureBio)Science60.568.2+7.7
APEX-Agents 1.1Professional Work79.471.8−7.6
GDP.pdf (Artificial Analysis, All-pass)Professional Work17.224.6+7.4
Humanity's Sixth SenseMultimodal30.122.7−7.4
LMArena Agent Arena Net Improvement (Steering Burden)Agents2.0-1.8−7.3
Humanity's Last ExamGeneral QA42.635.9−6.7
Finance Agent v2 - General Quantitative AnalysisFinance & Accounting79.272.7−6.5
CursorBench 4.0Code47.741.3−6.4
Finance Agent v2 - General Qualitative AnalysisFinance & Accounting74.167.8−6.3
CritPt (avg@5)Science24.030.0+6.0
HealthBench Professional (Length-Adjusted)Health & Medicine51.757.7+6.0
ARC-AGI-2General QA78.183.9+5.8
ALE-BenchCode20122154+5.0
LAB-Bench ProtocolQA (SecureBio)Science81.977.0−4.9
LMArena Agent Arena Confirmed SuccessAgents2.3-7.2−4.8
Concept Synth INDUCTION Challenge64 Round 1General QA32.837.5+4.7
ReactBench (39 tasks)Code34.439.0+4.6
FrontierCode 2.0 (Artificial Analysis)Code28.523.9−4.6
Excel Modeling BenchmarkFinance & Accounting61.866.2+4.4
VCT (SecureBio)Science42.947.2+4.3
Bio Capabilities Index (SecureBio)Indexes149155+4.2
LiveBench 2.0Agents79.275.0−4.2
Wiz Cyber Model Arena v3.9Cybersecurity60.364.4+4.1
NoCoT-Bench NCRI 15.2General QA81.689.0+4.1
Concept Synth INDUCTION Challenge100 Round 1General QA21.025.0+4.0
ZeroBench (Main questions, pass@5)Multimodal23.019.0−4.0
Legal Research BenchLegal45.241.3−3.8
AA-LCR v1.1Search82.779.0−3.7
Code MigrationCode44.247.8+3.6
Vals Index v2.1Indexes56.653.1−3.5
VCT v2 (SecureBio)Science38.541.9+3.4
Artificial Analysis Intelligence Index v4.3.2Indexes45.142.1−3.0
MBCT (SecureBio)Science53.956.9+3.0
RedlineBenchLegal44.947.8+2.9
τ³-Banking (Mercor run, bm25_grep retrieval)Agents41.638.8−2.8
GPQA-Bio (SecureBio)Science87.885.3−2.5
Finance Agent v2 - All PassFinance & Accounting45.242.8−2.5
DAYJOB: Healthcare (October 2026 rescoring)Health & Medicine6.03.6−2.4
ZeroBench Main Mean pass@1 (five samples)Multimodal11.814.2+2.4
AA-Omniscience IndexIndexes4.70.1−2.3
LAB-Bench CloningScenarios (SecureBio)Science94.892.5−2.3
LMArena Agent Arena Steering BurdenAgents4.0-0.5−2.3
LAB-Bench LitQA2 (SecureBio)Science49.051.1+2.1
Finance Agent v2 - Market AnalysisFinance & Accounting62.760.6−2.1
Political Science LLM BenchmarkProfessional Work66.268.3+2.0
Ruby LLM Benchmarks — Program FixerCode72.074.0+2.0
ARC-AGI-1 Semi-PrivateGeneral QA94.596.5+2.0
Finance Agent v2 - Earnings AnalysisFinance & Accounting75.377.3+2.0
WMDP-Bio (SecureBio)Science87.989.7+1.8
Finance Agent v2 - PrecedentsFinance & Accounting26.524.8−1.6
LMArena Agent Arena Praise vs ComplaintAgents1.2-1.9−1.6
Terminal-Bench 4.0 (Artificial Analysis)Code33.835.4+1.5
NoCoT-Bench NCKI (kspine v3)General QA102104+1.5
Epoch Capabilities Index (ECI)Indexes159161+1.2
APEX-1 (Mercor, mean score)Professional Work68.369.5+1.2
Chess Puzzles (task version 2.0)Games46.047.0+1.0
Terminal-Bench-Science 0.1Science12.913.8+1.0
SciCode 1.0.1 (Artificial Analysis)Code55.955.0−0.9
LMArena Agent Arena Bash ReliabilityAgents2.40.7−0.8
Finance Agent v2 - Disclosure AnalysisFinance & Accounting66.366.9+0.7
ARC-AGI-3 (Semi-Private)General QA0.20.8+0.6
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine26.727.2+0.6
Finance Agent v2 - ComparablesFinance & Accounting47.547.0−0.5
APEX-Accounting 1.1Finance & Accounting59.359.8+0.5
PubMedQA (SecureBio)Health & Medicine80.080.4+0.4
LMArena Agent Arena Tool HallucinationAgents0.31.0+0.4
Finance Agent v2 - Financial ModelingFinance & Accounting22.823.0+0.3
OTIS Mock AIME 2024-2025Math99.499.7+0.3
Artificial Analysis Intelligence Index v5.0.1Indexes32.632.3−0.2
MMMU ProMultimodal79.779.5−0.2
DeepsecBench-AA (Artificial Analysis)Cybersecurity24.925.1+0.1
Artificial Analysis Progress Index v1Indexes3243240.0
Harvey's Legal Agent Benchmark - Antitrust CompetitionLegal0.00.00.0
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 - Capital MarketsLegal0.00.00.0
Harvey's Legal Agent Benchmark - Corporate GovernanceLegal0.00.00.0
Harvey's Legal Agent Benchmark - Data Privacy CybersecurityLegal0.00.00.0
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 - InsuranceLegal0.00.00.0
Harvey's Legal Agent Benchmark - International Trade SanctionsLegal0.00.00.0
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
MMLU-Bio (SecureBio)Science90.790.70.0

DeepSeek V4.1 Flash 1017vsGPT-5.6 Terranormalised score marginsData Nov 15, 2026Data: Vals AI, Artificial Analysis, SecureBio and 32 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…