Atlas

Compare Models

DeepSeek V4.1 Flash 1017 vs Muse Wonder

Muse Wonder leads by 2.6 capability points (≈95% interval +0.2 to +5.0): a clear lead over DeepSeek V4.1 Flash 1017 · P(higher) ≈ 98%

Wins 96 of 121 shared benchmarks (24 lost, 1 tied)

Capability vs. relative cost
DeepSeek V4.1 Flash 1017Muse WonderClosed 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 1017vsMuse Wondercapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins24 – 96 · 121 shared, 1 tied

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

DeepSeek V4.1 Flash 1017vsMuse Wonderwins by categoryData Nov 15, 2026Data: Artificial Analysis, Vals AI, Surge AI and 30 othersVillage LLM Atlasatlas.village.dev

Margins

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

AA-PersonalityBench Warmth indexChat & Writing27.7100+73.1
HardGeoBench Avg DistanceMultimodal32011369+56.5
AA-PersonalityBench Imagination indexChat & Writing37.290.5+53.8
HardGeoBench Country AccuracyMultimodal15.052.5+37.5
AA-PersonalityBench Expressiveness indexChat & Writing52.488.6+36.5
HardGeoBench <=100miMultimodal7.538.8+31.3
AA-PersonalityBench Transparency indexChat & Writing86.756.2−30.8
DAYJOB: LawLegal6.236.6+30.4
DAYJOB: Finance (October 2026 rescoring)Finance & Accounting12.538.8+26.3
DAYJOB: Healthcare (October 2026 rescoring)Health & Medicine6.031.2+25.2
Visa Agentic Commerce BenchmarkAgents57.882.2+24.4
FrontierMath Tier 4 (v2) - PrivateMath46.370.7+24.4
HardGeoBench <=10miMultimodal3.827.5+23.8
DAYJOB: AccountingFinance & Accounting7.630.8+23.2
AA-PersonalityBench Informativeness indexChat & Writing71.494.3+23.1
BuseyBench SVGMultimodal4.36.6+22.5
LisanBench Path LengthGames2110727973+21.3
RiemannBench (November 2026 rescoring)Math38.459.2+20.8
CritPt v2 (agentic, avg@5)Science52.370.9+18.6
Code MigrationCode44.262.6+18.3
Humanity's Last Exam DiamondGeneral QA20.839.0+18.2
Terminal-Bench 4.0Code31.849.5+17.7
AA-PersonalityBench Conviction indexChat & Writing94.379.1−15.4
AA-PersonalityBench Slop indexChat & Writing39.154.3+15.4
VIALS-AA (pass^3)Multimodal40.455.3+14.9
Artificial Analysis Intelligence Index v5.0Indexes28.543.3+14.8
Humanity's Sixth SenseMultimodal30.143.7+13.6
DAYJOB: Data ScienceProfessional Work3.316.7+13.4
GDP.pdf (Artificial Analysis, All-pass)Professional Work17.230.6+13.4
Terminal-Bench-Science 0.1Science12.925.7+12.9
Artificial Analysis Engineering IndexIndexes37.649.6+12.0
VIALS-AA (average of 3)Multimodal53.264.8+11.6
Artificial Analysis Economics IndexIndexes38.549.5+11.0
Artificial Analysis Intelligence Index v5.0.1Indexes32.643.6+11.0
FrontierMath Tiers 1-3 (v2) - PrivateMath71.282.1+10.9
AA-Briefcase v1.1 EloAgents14161610+10.6
Bug Hunt BenchCode40.730.2−10.5
AA-Omniscience IndexIndexes4.725.5+10.4
τ³-Banking (Mercor run, bm25_grep retrieval)Agents41.651.9+10.3
Terminal-Bench 4.0 (Artificial Analysis)Code28.838.4+9.6
HLAB-Vals-1.1Legal14.223.8+9.6
Artificial Analysis Legal IndexIndexes37.746.9+9.1
Artificial Analysis Finance & Accounting IndexIndexes42.251.1+9.0
MKA (Artificial Analysis)Professional Work38.547.5+8.9
NoCoT-Bench NCRI 15.2General QA81.697.7+8.9
CWE-bench v2 Pass@1Cybersecurity57.666.1+8.5
VulnPR-100Code54.046.0−8.0
Harvey's Legal Agent BenchmarkLegal9.217.1+7.9
GameBench 2Games69.977.8+7.9
APEX-Accounting 1.1Finance & Accounting59.367.1+7.8
Artificial Analysis Coding Agent Index v2.1Indexes36.844.5+7.6
VIALS-AA (pass@3)Multimodal66.573.9+7.5
Artificial Analysis Healthcare & Medical IndexIndexes41.348.7+7.4
Artificial Analysis Intelligence Index v4.3.2Indexes45.152.4+7.3
APEX-1 (Mercor, mean score)Professional Work68.375.1+6.8
DAXBench v2Code77.784.0+6.4
LisanBench Difficulty-WeightedGames77338303+6.1
WeirdML v3Science9.215.2+6.0
FrontierCode 2.0 - Pass RateCode23.028.8+5.8
Finance Agent v2 - AdjustmentsFinance & Accounting58.452.7−5.7
Finance Agent v2 - PrecedentsFinance & Accounting26.532.1+5.6
APEX-SWECode49.154.6+5.5
HANDBOOK.md Agents v1.1Professional Work24.029.5+5.5
NoCoT-Bench NCKI (kspine v3)General QA102109+5.4
Legal Research BenchLegal45.250.5+5.3
Excel Modeling BenchmarkFinance & Accounting61.867.1+5.3
Artificial Analysis Strategy & Ops IndexIndexes49.254.5+5.3
MMMU ProMultimodal79.784.9+5.2
FrontierCode 2.0 - ScoreCode20.225.3+5.1
Humanity's Last ExamGeneral QA42.647.6+5.0
Code Migration v1.1Code67.572.4+4.8
DAYJOB: TaxFinance & Accounting1.26.0+4.8
Vibe Code Bench v1.1Code87.783.1−4.6
HoneyBench v0.2 (hack rate)Safety76.481.0−4.6
Vals Index v2.1Indexes56.661.1+4.5
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine26.722.2−4.4
SciCode 1.0.1 (Artificial Analysis)Code55.960.3+4.4
CritPt (avg@5)Science24.028.3+4.3
GBENCH One-shot Coding GScoreGames53.858.1+4.3
Artificial Analysis Progress Index v1Indexes324336+4.2
FrontierML ($200 wallet)Agents23.519.3−4.2
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Legal18.823.0+4.2
Tax Agent BenchFinance & Accounting74.971.2−3.7
STEMBench120 v1.0.0 (closed-book)Science2.52.7+3.6
Vals Index v2.2Indexes61.364.9+3.6
Vals Legal IndexIndexes71.975.3+3.4
Excel Modeling Benchmark v1.1Finance & Accounting69.272.5+3.3
Finance Agent v2 - General Qualitative AnalysisFinance & Accounting74.170.8−3.3
EnterpriseBench: CoreCraftProfessional Work58.561.5+3.0
FrontierCode 1.1 Main (Artificial Analysis)Code43.746.5+2.8
CursorBench 4.0Code47.750.4+2.7
Finance Agent v2.1Finance & Accounting62.965.5+2.6
Agents' Last Exam v1.1 (ALE v1.1)Agents36.138.5+2.5
APEX-Agents 1.1Professional Work79.477.0−2.4
Epoch Capabilities Index (ECI)Indexes159162+2.2
Mystery Game PuzzlesGames49.051.0+2.0
LMArena Code Arena WebDevCode16641649−2.0
AA-PersonalityBench Organization indexChat & Writing88.690.5+1.9
DeepSWE v1.1Code77.078.8+1.8
Finance Agent v2 - General Quantitative AnalysisFinance & Accounting79.277.6−1.6
LMArena Agent Arena Confirmed SuccessAgents2.35.4+1.5
LMArena Agent Arena Praise vs ComplaintAgents1.2-1.8−1.5
Finance Agent v2 - Earnings AnalysisFinance & Accounting75.376.8+1.5
GDPxProfessional Work5.46.8+1.4
LMArena Agent Arena Bash ReliabilityAgents2.44.8+1.2
LiveBench 2.0Agents79.278.1−1.1
DualEntry Accounting AI BenchmarkFinance & Accounting82.283.2+1.0
Finance Agent v2 - Disclosure AnalysisFinance & Accounting66.367.3+1.0
LMArena Agent Arena Steering BurdenAgents4.02.6−0.7
Finance Agent v2 - All PassFinance & Accounting45.245.9+0.7
Finance Agent v2 - Market AnalysisFinance & Accounting62.762.1−0.6
LMArena Agent Arena Net Improvement (Steering Burden)Agents2.01.8−0.4
AutomationBench-AA 1.0.6Agents68.568.1−0.4
LMArena Agent Arena Tool HallucinationAgents0.3-0.5−0.4
Finance Agent v2 - ComparablesFinance & Accounting47.547.9+0.4
Vals Finance IndexIndexes60.961.2+0.3
ErdosBenchMath3.03.0+0.3
Finance Agent v2 - Financial ModelingFinance & Accounting22.822.5−0.2
Build-A-Chip-Bench (20-task index run)Agents0.50.3−0.2
Access to Justice BenchmarkLegal73.173.0−0.1
OTIS Mock AIME 2024-2025Math99.499.40.0

DeepSeek V4.1 Flash 1017vsMuse Wondernormalised score marginsData Nov 15, 2026Data: Artificial Analysis, Vals AI, Surge AI and 30 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…