Provider guide
Where to run Qwen3.5-122B-A10B
6 live listings tracked — output prices vary 1.5× between the cheapest and most expensive host, so the provider choice matters as much as the model choice.
FASTEST MEASURED
65 tok/s
measured throughput · $0.26 input / $2.08 output per million tokens
Alibaba CloudCHEAPEST$0.26 in/1M$2.08 out/1M65 tok/s262K—2 hours agoOpenRouter$0.26 in/1M$2.08 out/1M—262K—2 hours agoSiliconFlowzero-retention$0.26 in/1M$2.08 out/1M40 tok/s262Kfp82 hours agoAtlasCloud$0.30 in/1M$2.40 out/1M26 tok/s262Kfp82 hours agoDeepInfra$0.29 in/1M$2.40 out/1M—262Kfp47 days agoNovita AIzero-retention through OpenRouterDirect and through OpenRouter$0.40 in/1M$3.20 out/1M40 tok/sthrough OpenRouter262Kbf162 hours ago
OpenRouter
DeepInfra
Novita AIFull specs, hardware verdicts and benchmarks on the Qwen3.5-122B-A10B model page →