Provider guide
Where to run Qwen3.5-35B-A3B
13 live offers tracked — output prices vary 2.0× between the cheapest and most expensive host, so the provider choice matters as much as the model choice.
FASTEST MEASURED
214 tok/s
measured throughput · $0.25 input / $1.25 output per million tokens
OpenRouterCHEAPEST$0.14 in/1M$1.00 out/1M—262K—17m agoDeepInfra$0.14 in/1M$1.00 out/1M—262Kfp86h agoAlibaba Cloud$0.16 in/1M$1.30 out/1M159 tok/s262Kfp815m agoAlibaba Cloud$0.16 in/1M$1.30 out/1M146 tok/s262K—6d agoAtlasCloud$0.23 in/1M$1.80 out/1M90 tok/s262Kfp815m agoNovita AI$0.25 in/1M$2.00 out/1M—262K—6h agoDeepInfrazero-retention$0.14 in/1M$1.00 out/1M81 tok/s262Kfp815m agoParasailzero-retention$0.15 in/1M$1.00 out/1M26 tok/s262Kfp815m agoAkashMLzero-retention$0.14 in/1M$1.00 out/1M200 tok/s262Kfp815m agoVenice AIzero-retention$0.31 in/1M$1.25 out/1M141 tok/s256K—15m agoCoreWeavezero-retention$0.25 in/1M$1.25 out/1M214 tok/s262Kfp815m agoNextBitzero-retention$0.22 in/1M$1.40 out/1M45 tok/s262Kfp815m agoSiliconFlowzero-retention$0.24 in/1M$1.80 out/1M76 tok/s262Kfp815m ago
DeepInfra
Novita AIFull specs, hardware verdicts and benchmarks on theQwen3.5-35B-A3B model page →