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 Indexagentic8.3mirroropenrouterSep 20, 2026
8.3independentaaSep 20, 2026
AA Coding Indexcoding48.2mirroropenrouterJul 23, 2026
48.2independentaaJul 24, 2026
AA Intelligence Indexintelligence21.4non-reasoningindependentaaSep 9, 2026
18.4mirroropenrouterSep 20, 2026
18.4independentaaSep 20, 2026
Design Arena — 3dpreference1190mirroropenrouterSep 20, 2026
Design Arena — codecategoriespreference1193mirroropenrouterSep 21, 2026
Design Arena — datavizpreference1190mirroropenrouterSep 17, 2026
Design Arena — gamedevpreference1164mirroropenrouterSep 20, 2026
Design Arena — svgpreference1162mirroropenrouterSep 21, 2026
Design Arena — uicomponentpreference1179mirroropenrouterSep 20, 2026
Design Arena — websitepreference1203mirroropenrouterSep 11, 2026
LMArena Textpreference1438crowdlmarenaSep 13, 2026
LMArena Visionpreference1265crowdlmarenaSep 13, 2026
LMArena WebDevpreference1399crowdlmarenaSep 11, 2026
Median output speedspeed89 tok/sindependentaaSep 21, 2026
91 tok/snon-reasoningindependentaaSep 21, 2026
Median time to first tokenspeed2.17snon-reasoningindependentaaSep 20, 2026
2.22sindependentaaSep 21, 2026

Includes index data from Artificial Analysis.

Providers

Live operational stats per hosting endpoint, via OpenRouter.

ProviderQuantContext$/1M in · outUptime 24hLatencyThroughput
Alibaba262K$0.39 · $2.3499.96%
AtlasCloudfp8262K$0.55 · $3.596.15%
DeepInfrafp8262K$0.45 · $399.85%
DigitalOcean131K$0.55 · $3.591.59%
GMICloudfp8262K$0.6 · $3.684.46%
Novita262K$0.6 · $3.698.85%
Parasailfp8262K$0.5 · $3.699.78%
Phala262K$0.55 · $3.597.62%
StreamLake256K$0.6 · $3.698.21%
Venice128K$0.75 · $4.570.54%