llm.ing

Qwen3.5 122B-A10B

Alibabaopen weightsOpenRouter ↗Hugging Face ↗Artificial Analysis ↗

Qwen vision-language model for visual reasoning, documents, and agent tasks

Benchmarks

BenchmarkCategoryScoreVariantProvenanceSourceObserved
AA Agentic Indexagentic16.2non-reasoningindependentaaAug 7, 2026
7.6—mirroropenrouterSep 20, 2026
7.6—independentaaSep 20, 2026
AA Coding Indexcoding43.3non-reasoningindependentaaJul 24, 2026
45.7—independentaaJul 24, 2026
45.7—mirroropenrouterOct 28, 2026
AA Intelligence Indexintelligence17.7non-reasoningindependentaaSep 9, 2026
15.6—independentaaSep 20, 2026
17.7—mirroropenrouterOct 5, 2026
LMArena Textpreference1417—crowdlmarenaNov 9, 2026
LMArena Visionpreference1244—crowdlmarenaNov 9, 2026
LMArena WebDevpreference1360—crowdlmarenaNov 11, 2026
Median output speedspeed127 tok/s—independentaaNov 18, 2026
145 tok/snon-reasoningindependentaaNov 19, 2026
Median time to first tokenspeed2.26snon-reasoningindependentaaNov 19, 2026
2.24s—independentaaNov 19, 2026

Includes index data from Artificial Analysis.

Providers

Live operational stats per hosting endpoint, via OpenRouter.

ProviderQuantContext$/1M in · outUptime 24hLatencyThroughput
Alibaba—262K$0.26 · $2.0899.78%——
AtlasCloudfp8262K$0.3 · $2.477.23%——
Novitabf16262K$0.4 · $3.2100.00%——
SiliconFlowfp8262K$0.26 · $2.08100.00%——