Provider guide
Where to run Qwen3 VL 30B A3B Thinking
4 live listings tracked — output prices vary 2.4× between the cheapest and most expensive host, so the provider choice matters as much as the model choice.
FASTEST MEASURED
113 tok/s
measured throughput · $0.20 input / $2.40 output per million tokens
Novita AI$0.20 in/1M$1.00 out/1M—131K—21 days agoSiliconFlowzero-retention$0.29 in/1M$1.00 out/1M83 tok/s262Kfp83 hours agoOpenRouterCHEAPEST$0.20 in/1M$2.40 out/1M—262K—3 hours agoAlibaba Cloud$0.20 in/1M$2.40 out/1M113 tok/s131K—3 hours ago
Novita AIFull specs, hardware verdicts and benchmarks on the Qwen3 VL 30B A3B Thinking model page →