llm.ing

Qwen3.5 397B-A17B

Alibabaopen weightsOpenRouter ↗Hugging Face ↗Artificial Analysis ↗

Large open Qwen multimodal MoE for visual agents and long technical tasks

Benchmarks

BenchmarkCategoryScoreVariantProvenanceSourceObserved
AA Agentic Indexagentic8.3—mirroropenrouterSep 20, 2026
8.3—independentaaSep 20, 2026
AA Coding Indexcoding48.2—mirroropenrouterJul 23, 2026
48.2—independentaaJul 24, 2026
AA Intelligence Indexintelligence21.4non-reasoningindependentaaSep 9, 2026
18.4—independentaaSep 20, 2026
21.4—mirroropenrouterOct 5, 2026
Design Arena — 3dpreference1128—mirroropenrouterNov 20, 2026
Design Arena — codecategoriespreference1163—mirroropenrouterNov 20, 2026
Design Arena — datavizpreference1171—mirroropenrouterNov 19, 2026
Design Arena — gamedevpreference1120—mirroropenrouterNov 20, 2026
Design Arena — svgpreference1124—mirroropenrouterNov 19, 2026
Design Arena — uicomponentpreference1148—mirroropenrouterNov 20, 2026
Design Arena — websitepreference1178—mirroropenrouterNov 20, 2026
LMArena Textpreference1438—crowdlmarenaNov 18, 2026
LMArena Visionpreference1262—crowdlmarenaNov 9, 2026
LMArena WebDevpreference1400—crowdlmarenaNov 18, 2026
Median output speedspeed58 tok/s—independentaaNov 20, 2026
58 tok/snon-reasoningindependentaaNov 20, 2026
Median time to first tokenspeed2.11s—independentaaNov 20, 2026
2.12snon-reasoningindependentaaNov 20, 2026

Includes index data from Artificial Analysis.

Providers

Live operational stats per hosting endpoint, via OpenRouter.

ProviderQuantContext$/1M in · outUptime 24hLatencyThroughput
Alibaba—262K$0.39 · $2.34100.00%——
AtlasCloudfp8262K$0.55 · $3.5100.00%——
DeepInfrafp8262K$0.45 · $396.28%——
DeepInfrafp8262K$0.36 · $2.4100.00%——
DeepInfrafp8262K$0.675 · $4.579.93%——
GMICloudfp8262K$0.6 · $3.6100.00%——
Novita—262K$0.6 · $3.6100.00%——
Parasailfp8262K$0.5 · $3.698.90%——
StreamLake—256K$0.6 · $3.697.33%——
Venice—128K$0.75 · $4.597.55%——