Atlas

Compare Models

Claude Opus 5 vs DeepSeek V4.1 Flash 1017

Claude Opus 5 leads by 3.1 capability points (≈95% interval +0.7 to +5.6): a clear lead over DeepSeek V4.1 Flash 1017 · P(higher) ≈ 99%

Wins 86 of 130 shared benchmarks (25 lost, 19 tied)

Capability vs. relative cost
Claude Opus 5DeepSeek V4.1 Flash 1017Closed 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.

Claude Opus 5vsDeepSeek V4.1 Flash 1017capability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins86 – 25 · 130 shared, 19 tied

Comparing matching setups only (9 benchmarks excluded) · suites counted once
← Claude Opus 5DeepSeek V4.1 Flash 1017 →

Claude Opus 5vsDeepSeek V4.1 Flash 1017wins by categoryData Nov 15, 2026Data: Vals AI, Artificial Analysis, SecureBio and 38 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is DeepSeek V4.1 Flash 1017 minus Claude Opus 5.

HardGeoBench Avg DistanceMultimodal12183201−61.2
HardGeoBench <=100miMultimodal46.37.5−38.7
HardGeoBench Country AccuracyMultimodal53.815.0−38.7
HardGeoBench <=10miMultimodal32.53.8−28.7
LisanBench Difficulty-WeightedGames50687733+28.5
FrontierMath Tier 4 (v2) - PrivateMath73.246.3−26.8
ARC-AGI-3 (Semi-Private)General QA30.25.8−24.4
Humanity's Last Exam DiamondGeneral QA44.820.8−24.0
Terminal-Bench 4.0Code53.931.0−22.9
ALE-BenchCode26222012−21.6
Bio Capabilities Index (SecureBio)Indexes178149−21.3
CritPt v2 (agentic, avg@5)Science71.450.6−20.9
Harvey's Legal Agent Benchmark - Emerging Companies Venture CapitalLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Funds Asset ManagementLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - ImmigrationLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Intellectual PropertyLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Litigation Dispute ResolutionLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Structured Finance SecuritizationLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - Trusts Estates Private ClientLegal0.020.0+20.0
BuseyBench SVGMultimodal6.34.3−19.5
WCB (SecureBio)Science68.849.6−19.2
NoCoT-Bench NCRI 15.2General QA11581.6−18.4
HPCT (SecureBio)Science78.660.5−18.1
LisanBench Path LengthGames1542821107+17.6
AA-Omniscience IndexIndexes37.14.7−16.2
DAYJOB: Healthcare (October 2026 rescoring)Health & Medicine22.06.0−16.0
Terminal-Bench-Science 0.1Science28.612.9−15.7
Terminal-Bench 4.0 (Artificial Analysis)Code49.033.8−15.2
GameBench 2Games84.869.9−14.9
APEX-SWECode63.749.1−14.6
FrontierMath Tiers 1-3 (v2) - PrivateMath85.671.2−14.4
GBENCH One-shot Coding GScoreGames68.153.8−14.4
Humanity's Last ExamGeneral QA56.442.6−13.8
Code MigrationCode57.544.2−13.3
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine13.926.7+12.8
ARC-AGI-2General QA90.478.1−12.4
NoCoT-Bench NCKI (kspine v3)General QA119102−12.3
Arena Image-to-WebDevCode16641609−11.9
AutomationBench-AA 1.0.6Agents56.668.5+11.9
Excel Modeling BenchmarkFinance & Accounting73.661.8−11.8
Artificial Analysis Intelligence Index v5.0.1Indexes44.232.6−11.6
Bug Hunt BenchCode25.737.3+11.6
DAYJOB: Finance (October 2026 rescoring)Finance & Accounting23.812.5−11.3
Wiz Cyber Model Arena v3.9Cybersecurity71.260.3−10.9
AutomationBench 1.0.6 (provider-reported)Agents50.360.5+10.2
EnterpriseBench: CoreCraftProfessional Work68.758.5−10.2
Legal Research BenchLegal55.345.2−10.1
Mystery Game PuzzlesGames59.049.0−10.0
LMArena Agent Arena Net Improvement (Steering Burden)Agents7.22.0−9.9
Bedside BenchHealth & Medicine94.685.7−8.9
DAXBench v2Code86.277.7−8.5
VCT (SecureBio)Science51.442.9−8.5
FrontierCode 2.0 - Pass RateCode31.423.0−8.4
FrontierCode 2.0 - ScoreCode28.120.2−7.9
ProgramBench - almost solved rollout percentageCode37.029.1−7.9
Ruby LLM Benchmarks — Program FixerCode79.872.0−7.9
AA-Briefcase v1.1 EloAgents15581416−7.8
ReactBench (39 tasks)Code42.134.4−7.7
BrowseCompSearch90.883.4−7.4
NL2Repo - release-reported repository generation scoreCode75.368.2−7.1
Agents' Last Exam - release-reported pass rate, subset unspecifiedAgents28.635.4+6.8
MBCT (SecureBio)Science60.253.9−6.3
VulnPR-100Code48.054.0+6.0
FrontierPhysics v0.1Science9.53.6−5.9
Artificial Analysis Intelligence Index v4.3.2Indexes50.845.1−5.7
LMArena Agent Arena Praise vs ComplaintAgents12.21.2−5.5
ZeroBench Main Mean pass@1 (five samples)Multimodal17.211.8−5.4
Design Arena Overall Code CategoriesCode13211289−5.3
Political Science LLM BenchmarkProfessional Work71.466.2−5.2
LAB-Bench CloningScenarios (SecureBio)Science100.094.8−5.2
CritPt (avg@5)Science29.124.0−5.1
GPQA-Bio (SecureBio)Science82.787.8+5.1
APEX-Accounting 1.1Finance & Accounting64.359.3−5.0
GDP.pdf (Artificial Analysis, All-pass)Professional Work21.617.2−4.4
MMMU ProMultimodal84.079.7−4.3
Epoch Capabilities Index (ECI)Indexes164159−4.2
LMArena Agent Arena Confirmed SuccessAgents10.22.3−3.9
DeepSWE v1.1Code73.277.0+3.8
LMArena Agent Arena Bash ReliabilityAgents9.82.4−3.7
CharXiv Reasoning (with tools)Multimodal89.385.6−3.7
LMArena Code Arena WebDevCode16911664−3.6
Artificial Analysis Progress Index v1Indexes334324−3.5
ARC-AGI-1 Semi-PrivateGeneral QA97.594.5−3.0
Chess Puzzles (task version 2.0)Games43.046.0+3.0
Concept Synth INDUCTION Challenge100 Round 1General QA24.021.0−3.0
ZeroBench (Main questions, pass@5)Multimodal26.023.0−3.0
LiveBench 2.0Agents82.179.2−2.9
APEX-Agents 1.1Professional Work77.079.4+2.4
LAB-Bench LitQA2 (SecureBio)Science51.449.0−2.4
τ³-Banking (Mercor run, bm25_grep retrieval)Agents44.041.6−2.4
DeepsecBench — Revalidated V1 (233 issues)Code32.330.0−2.3
τ³-Banking (Fireworks, banking_knowledge)Agents49.146.8−2.3
Harvey HLAB — Criteria Pass RateLegal87.789.7+2.0
Tax Agent BenchFinance & Accounting76.774.9−1.7
BabyVision with toolsMultimodal94.192.5−1.6
Epiq LAER CorporateBenchProfessional Work74.075.2+1.2
CursorBench 4.0Code46.647.7+1.1
Chartography (with tools)Multimodal84.083.0−1.0
ZeroBench (Main questions, pass@5, with tools)Multimodal52.051.0−1.0
DeepsecBench-AA (Artificial Analysis)Cybersecurity25.924.9−0.9
RedlineBenchLegal44.044.9+0.9
LAB-Bench ProtocolQA (SecureBio)Science81.181.9+0.8
Vibe Code Bench v1.1Code88.487.7−0.7
AA-LCR v1.1Search82.082.7+0.7
OTIS Mock AIME 2024-2025Math98.999.4+0.6
SciCode 1.0.1 (Artificial Analysis)Code56.455.9−0.5
Humanity's Sixth SenseMultimodal30.530.1−0.4
MMLU-Bio (SecureBio)Science91.190.7−0.4
LMArena Agent Arena Tool HallucinationAgents0.90.3−0.3
APEX-1 (Mercor, mean score)Professional Work68.668.3−0.3
LMArena Agent Arena Steering BurdenAgents4.44.0−0.2
Aikido Known-CVE Detection (32 CVEs) single-run averageCybersecurity70.870.80.0
Concept Synth INDUCTION Challenge64 Round 1General QA32.832.80.0
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 - Corporate M&ALegal20.020.00.0
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 - 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 - 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

Claude Opus 5vsDeepSeek V4.1 Flash 1017normalised score marginsData Nov 15, 2026Data: Vals AI, Artificial Analysis, SecureBio and 38 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…