Atlas

Compare Models

GPT-5.5 vs Qwen3.8 Max (0902)

Qwen3.8 Max (0902) leads by 0.4 capability points (≈95% interval −1.9 to +2.6): statistically tied · P(higher) ≈ 62%

Wins 18 of 30 shared benchmarks (12 lost)

Capability vs. relative cost
GPT-5.5Qwen3.8 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.

GPT-5.5vsQwen3.8 Max (0902)capability vs. relative costData Nov 17, 2026Village LLM Atlasatlas.village.dev

Benchmark wins12 – 18 · 30 shared

Comparing matching setups only (74 benchmarks excluded) · suites counted once
← GPT-5.5Qwen3.8 Max (0902) →

GPT-5.5vsQwen3.8 Max (0902)wins by categoryData Nov 17, 2026Data: Artificial Analysis, Vals AI, BuseyBench and 4 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 GPT-5.5.

AA-Briefcase EloAgents11581622+27.0
Terminal-Bench 4.0 (Artificial Analysis)Code14.638.9+24.2
Vibe Code Bench v1.1Code58.277.1+18.9
GDPval-AA v2Professional Work13721689+15.6
GDPval-AA v2.1Professional Work13571674+14.0
AutomationBench-AA 1.0.6Agents47.356.2+8.9
𝜏³-BankingAgents39.047.8+8.9
CritPt (avg@5)Science25.417.7−7.7
Artificial Analysis Intelligence Index v4.3.2Indexes38.445.4+7.1
Artificial Analysis Intelligence Index v4.3Indexes38.645.4+6.8
Finance Agent v2Finance & Accounting51.857.9+6.1
ITBench-AAAgents45.840.3−5.6
DAXBench v2Code88.383.0−5.3
AA-AnalystAgent (pass^5)Professional Work50.045.0−5.0
APEX-Agents-AA (avg@3)Professional Work37.742.4+4.7
Terminal-Bench 2.1Code84.388.8+4.5
AA-Omniscience IndexIndexes20.512.0−4.3
SciCode 1.0.1 (Artificial Analysis)Code56.152.1−4.1
AA-LCR v1.1Search84.380.3−4.0
Humanity's Last Exam DiamondGeneral QA27.924.0−3.9
Epoch Capabilities Index (ECI)Indexes160156−3.4
Medical Long Context Reasoning (MLCR-AA)Health & Medicine17.820.0+2.2
BuseyBench SVGMultimodal6.05.8−2.1
Humanity's Last Exam (Text-Only)General QA45.043.1−1.9
Code2Video-AA EloCode10001021+1.7
MMMU ProMultimodal81.282.8+1.6
Artificial Analysis Progress Index v1Indexes320324+1.3
EnterpriseOps-Gym-AAAgents46.647.6+1.0
GPQA Diamond (Artificial Analysis avg@5)General QA93.592.8−0.7
GDP.pdf (Artificial Analysis, All-pass)Professional Work22.622.8+0.2

GPT-5.5vsQwen3.8 Max (0902)normalised score marginsData Nov 17, 2026Data: Artificial Analysis, Vals AI, BuseyBench and 4 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…