Atlas

Compare Models

DeepSeek V4.1 Flash vs Qwen3.8 Max (0902)

Qwen3.8 Max (0902) leads by 0.9 capability points (≈95% interval −1.3 to +3.1): statistically tied · P(higher) ≈ 80%

Wins 20 of 34 shared benchmarks (14 lost)

Capability vs. relative cost
DeepSeek V4.1 FlashQwen3.8 Max (0902)Closed 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 FlashvsQwen3.8 Max (0902)capability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins14 – 20 · 34 shared

Comparing matching setups only (109 benchmarks excluded) · suites counted once
← DeepSeek V4.1 FlashQwen3.8 Max (0902) →

DeepSeek V4.1 FlashvsQwen3.8 Max (0902)wins by categoryData Nov 15, 2026Data: Artificial Analysis, BuseyBench, DAXBench and 5 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Qwen3.8 Max (0902) minus DeepSeek V4.1 Flash.

AA-PersonalityBench Conviction indexChat & Writing96.235.3−61.5
AA-PersonalityBench Informativeness indexChat & Writing79.139.1−40.4
AA-PersonalityBench Expressiveness indexChat & Writing61.923.8−38.5
AA-PersonalityBench Slop indexChat & Writing37.273.3+36.5
AA-PersonalityBench Organization indexChat & Writing10071.4−28.8
AA-PersonalityBench Imagination indexChat & Writing31.548.6+17.3
BuseyBench SVGMultimodal4.15.8+17.0
AA-PersonalityBench Transparency indexChat & Writing60.044.8−15.4
AutomationBench-AA 1.0.6Agents68.956.2−12.7
Terminal-Bench 4.0 (Artificial Analysis)Code26.838.9+12.1
AA-Briefcase EloAgents14241622+11.5
AA-AnalystAgent (pass^5)Professional Work35.045.0+10.0
GDP.pdf (Artificial Analysis, All-pass)Professional Work12.822.8+10.0
AA-Omniscience IndexIndexes-5.312.0+8.6
AA-PersonalityBench Warmth indexChat & Writing41.048.6+7.7
DAXBench v2Code75.583.0+7.4
ITBench-AAAgents46.940.3−6.7
Artificial Analysis Intelligence Index v4.3.2Indexes39.545.4+6.0
Artificial Analysis Intelligence Index v4.3Indexes39.545.4+5.9
SPU (Science Prediction Universe)Science0.20.1−5.5
Humanity's Last Exam DiamondGeneral QA19.124.0+4.9
TCSExamMath40.444.8+4.4
Terminal-Bench 4.0Code31.227.0−4.2
CritPt (avg@5)Science14.317.7+3.4
GDPval-AA v2.1Professional Work16001674+3.3
Terminal-Bench-Science 0.1Science9.011.9+2.9
GDPval-AA v2Professional Work16321689+2.8
Medical Long Context Reasoning (MLCR-AA)Health & Medicine22.820.0−2.8
Code2Video-AA EloCode10531021−2.6
Terminal-Bench 2.1Code90.688.8−1.8
Artificial Analysis Progress Index v1Indexes320324+1.2
FrontierCode 1.1 Main (Artificial Analysis)Code44.043.2−0.9
Epoch Capabilities Index (ECI)Indexes156156−0.5
SciCode 1.0.1 (Artificial Analysis)Code51.952.1+0.2

DeepSeek V4.1 FlashvsQwen3.8 Max (0902)normalised score marginsData Nov 15, 2026Data: Artificial Analysis, BuseyBench, DAXBench and 5 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…