Atlas

Compare Models

Gemini 3.8 Flash vs Qwen3.8-Max

Gemini 3.8 Flash leads by 1.0 capability points (≈95% interval −1.4 to +3.4): statistically tied · P(higher) ≈ 79%

Wins 27 of 34 shared benchmarks (7 lost)

Capability vs. relative cost
Gemini 3.8 FlashQwen3.8-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.

Gemini 3.8 FlashvsQwen3.8-Maxcapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins27 – 7 · 34 shared

Comparing matching setups only (293 benchmarks excluded) · suites counted once
← Gemini 3.8 FlashQwen3.8-Max →

Gemini 3.8 FlashvsQwen3.8-Maxwins by categoryData Nov 15, 2026Data: Artificial Analysis, Conceptual Reasoning Benchmark Authors, Alibaba and 11 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 Gemini 3.8 Flash.

BabaIsBench Stage 0: The IntroGames100.058.0−42.0
DeepSWE v1.1Code73.856.6−17.2
AA-Omniscience IndexIndexes29.63.4−13.1
GameBench 2Games64.051.4−12.6
BuseyBench SVGMultimodal6.65.4−12.1
AutomationBench-AA 1.0.6Agents60.949.2−11.7
AA-Briefcase EloAgents12021388+10.8
Conceptual Reasoning IndexIndexes65.759.4−8.7
GDPval-AA v2Professional Work14641630+8.2
GDPval-AA v2.1Professional Work14421611+7.5
Revelo Code IndexCode36.729.4−7.3
ACCoRDGeneral QA78.271.4−6.7
LMCAGeneral QA52.946.2−6.7
Terminal-Bench 2.1Code87.681.3−6.4
AA-LCR v1.1Search84.078.3−5.7
𝜏³-BankingAgents45.851.3+5.6
LVBenchMultimodal87.181.8−5.3
Ramp Accounting BenchFinance & Accounting45.741.2−4.6
Continual Learning Exam (Trial 1)Agents21.225.7+4.5
DTBenchGeneral QA95.792.0−3.7
SciCode 1.0.1 (Artificial Analysis)Code56.653.2−3.4
GDP.pdf (Artificial Analysis, All-pass)Professional Work22.820.2−2.6
GPQA Diamond (Artificial Analysis avg@5)General QA95.392.7−2.5
Medical Long Context Reasoning (MLCR-AA)Health & Medicine21.719.4−2.2
CharXiv ReasoningMultimodal86.288.4+2.2
MMMU ProMultimodal84.582.3−2.2
RSI-Exam 0.1 (Full 88)Agents0.30.3−1.9
CritPt (avg@5)Science18.320.0+1.7
Terminal-Bench 4.0 (Artificial Analysis)Code19.718.7−1.0
Epoch Capabilities Index (ECI)Indexes159157−1.0
Humanity's Last ExamGeneral QA44.543.6−0.9
Artificial Analysis Intelligence Index v4.3Indexes41.240.3−0.9
Artificial Analysis Intelligence Index v4.3.2Indexes40.940.2−0.8
Artificial Analysis Progress Index v1Indexes320320−0.2

Gemini 3.8 FlashvsQwen3.8-Maxnormalised score marginsData Nov 15, 2026Data: Artificial Analysis, Conceptual Reasoning Benchmark Authors, Alibaba and 11 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…