Atlas

Compare Models

DeepSeek V4.1 Flash 1017 vs Qwen3.8-Max

DeepSeek V4.1 Flash 1017 leads by 1.3 capability points (≈95% interval −1.3 to +3.8): statistically tied · P(higher) ≈ 83%

Wins 17 of 22 shared benchmarks (5 lost)

Capability vs. relative cost
DeepSeek V4.1 Flash 1017Qwen3.8-MaxClosed 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 1017vsQwen3.8-Maxcapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins17 – 5 · 22 shared

Comparing matching setups only (160 benchmarks excluded) · suites counted once
← DeepSeek V4.1 Flash 1017Qwen3.8-Max →

DeepSeek V4.1 Flash 1017vsQwen3.8-Maxwins by categoryData Nov 15, 2026Data: Artificial Analysis, DeepSeek, Alibaba and 6 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Qwen3.8-Max minus DeepSeek V4.1 Flash 1017.

AutomationBench-AA 1.0.6Agents68.549.2−19.3
GameBench 2Games69.951.4−18.5
Cresta-BenchAgents32.648.7+16.0
Terminal-Bench 4.0 (Artificial Analysis)Code33.818.7−15.2
DeepsecBench — Revalidated V1 (233 issues)Code30.015.5−14.5
NL2Repo - release-reported repository generation scoreCode68.255.9−12.3
BuseyBench SVGMultimodal4.35.4+10.7
Agents' Last Exam - release-reported pass rate, subset unspecifiedAgents35.427.0−8.4
Artificial Analysis Intelligence Index v4.3.2Indexes45.140.2−4.9
AA-LCR v1.1Search82.778.3−4.3
CritPt (avg@5)Science24.020.0−4.0
GDP.pdf (Artificial Analysis, All-pass)Professional Work17.220.2+3.0
SciCode 1.0.1 (Artificial Analysis)Code55.953.2−2.7
MMMU ProMultimodal79.782.3+2.6
ZeroBench (Main questions, pass@5, with tools)Multimodal51.049.0−2.0
AA-Briefcase v1.1 EloAgents14161382−1.9
Epoch Capabilities Index (ECI)Indexes159157−1.5
Artificial Analysis Progress Index v1Indexes324320−1.4
BabyVision with toolsMultimodal92.591.3−1.2
Humanity's Last ExamGeneral QA42.643.6+1.0
AA-Omniscience IndexIndexes4.73.4−0.6
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine26.726.1−0.6

DeepSeek V4.1 Flash 1017vsQwen3.8-Maxnormalised score marginsData Nov 15, 2026Data: Artificial Analysis, DeepSeek, Alibaba and 6 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…