Atlas

Compare Models

Gemini 3.7 Flash vs Qwen3.8-Max

Qwen3.8-Max leads by 0.3 capability points (≈95% interval −2.0 to +2.6): statistically tied · P(higher) ≈ 60%

Wins 20 of 37 shared benchmarks (17 lost)

Capability vs. relative cost
Gemini 3.7 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 3.7 FlashvsQwen3.8-Maxcapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins17 – 20 · 37 shared

Comparing matching setups only (278 benchmarks excluded) · suites counted once
← Gemini 3.7 FlashQwen3.8-Max →

Gemini 3.7 FlashvsQwen3.8-Maxwins by categoryData Nov 15, 2026Data: Artificial Analysis, Conceptual Reasoning Benchmark Authors, Alibaba and 10 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 3.7 Flash.

BabaIsBench Stage 0: The IntroGames100.058.0−42.0
𝜏³-BankingAgents35.551.3+15.9
AA-Briefcase EloAgents11161388+15.8
AutomationBench-AA 1.0.6Agents62.049.2−12.8
BuseyBench SVGMultimodal6.65.4−12.2
AA-Omniscience IndexIndexes26.53.4−11.5
GDPval-AA v2Professional Work14331630+9.7
GDPval-AA v2.1Professional Work14041611+9.1
BabyVisionMultimodal73.582.0+8.5
Conceptual Reasoning IndexIndexes64.659.4−7.1
SciCode 1.0.1 (Artificial Analysis)Code59.853.2−6.6
CritPt (avg@5)Science14.320.0+5.7
ACCoRDGeneral QA77.071.4−5.6
MRCR v2 (8-needle, Context Arena) - 128K pointwiseGeneral QA82.187.3+5.3
FrontierChallenge (Pass Rate)Science10.315.5+5.2
Terminal-Bench 4.0 (Artificial Analysis)Code13.618.7+5.1
DTBenchGeneral QA96.892.0−4.8
AA-LCR v1.1Search83.078.3−4.7
Terminal-Bench 2.1Code85.881.3−4.5
Medical Long Context Reasoning (MLCR-AA)Health & Medicine15.019.4+4.4
LMCAGeneral QA50.446.2−4.1
Humanity's Last Exam (Text-Only)General QA39.043.0+4.1
CharXiv ReasoningMultimodal84.588.4+3.9
MRCR v2 (8-needle, Context Arena) - 128K AUCGeneral QA89.593.3+3.9
LVBenchMultimodal85.481.8−3.6
GDP.pdf (Artificial Analysis, All-pass)Professional Work23.620.2−3.4
xbench-ScienceQAScience70.273.4+3.2
MMMU ProMultimodal84.982.3−2.5
GPQA Diamond (Artificial Analysis avg@5)General QA94.592.7−1.8
RSI-Exam 0.1 (Full 88)Agents0.30.3+1.3
Ramp Accounting BenchFinance & Accounting42.341.2−1.1
Artificial Analysis Intelligence Index v4.3.2Indexes39.140.2+1.1
Artificial Analysis Intelligence Index v4.3Indexes39.440.3+0.9
Epoch Capabilities Index (ECI)Indexes158157−0.8
Artificial Analysis Progress Index v1Indexes317320+0.8
Agents' Last Exam - release-reported pass rate, subset unspecifiedAgents26.327.0+0.7
Continual Learning Exam (Trial 1)Agents25.325.7+0.3

Gemini 3.7 FlashvsQwen3.8-Maxnormalised score marginsData Nov 15, 2026Data: Artificial Analysis, Conceptual Reasoning Benchmark Authors, Alibaba and 10 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…