Provider guide
Where to run Qwen3 Coder 480B A35B
6 live listings tracked — output prices vary 4.9× between the cheapest and most expensive host, so the provider choice matters as much as the model choice.
FASTEST MEASURED
72 tok/s
measured throughput · $0.30 input / $1.00 output per million tokens
OpenRouterCHEAPEST$0.30 in/1M$1.00 out/1M—262K—2 hours agoVenice AIzero-retention$0.35 in/1M$1.50 out/1M39 tok/s256Kfp82 hours agoNovita AIzero-retention through OpenRouterDirect and through OpenRouter$0.38 in/1M$1.55 out/1M42 tok/sthrough OpenRouter262Kfp88 hours agoGoogle Vertex AIzero-retention$0.22 in/1M$1.80 out/1M55 tok/s262K—2 hours agoAlibaba Cloud$0.97 in/1M$4.88 out/1M36 tok/s262K—2 hours agoDeepInfraturbo tier$0.30 in/1M$1.00 out/1M72 tok/s262Kfp42 hours ago
Novita AI
Google Vertex AIFull specs, hardware verdicts and benchmarks on the Qwen3 Coder 480B A35B model page →