Provider guide
Where to run Kimi K2.7 Code
20 live offers tracked — output prices vary 2.3× between the cheapest and most expensive host, so the provider choice matters as much as the model choice.
FASTEST MEASURED
215 tok/s
measured throughput · $0.95 input / $4.00 output per million tokens
OpenRouterCHEAPEST$0.73 in/1M$3.50 out/1M—262K—13m agoDeepInfra$0.74 in/1M$3.50 out/1M—262Kfp46h agoAmbient$0.73 in/1M$3.50 out/1M34 tok/s262Kint412m agoNovita AI$0.95 in/1M$4.00 out/1M—262K—6h agoAlibaba Cloud$0.95 in/1M$4.00 out/1M36 tok/s262Kfp812m agoCloudflare Workers AI$0.95 in/1M$4.00 out/1M60 tok/s262K—12m agoAtlasCloud$0.95 in/1M$4.00 out/1M52 tok/s262Kint412m agoNebius AI Studio$0.95 in/1M$4.00 out/1M177 tok/s8Kfp44d agoDeepInfrazero-retention$0.74 in/1M$3.50 out/1M62 tok/s262Kfp412m agoVenice AIzero-retention$0.75 in/1M$3.50 out/1M23 tok/s256Kint412m agoParasailzero-retention$0.76 in/1M$3.50 out/1M65 tok/s262Kint412m agoCoreWeavezero-retention$0.71 in/1M$3.50 out/1M197 tok/s262Kint412m agoInceptronzero-retention$0.72 in/1M$3.50 out/1M60 tok/s262Kint412m agoModelRunzero-retention$0.85 in/1M$3.75 out/1M176 tok/s262Kfp412m agoSiliconFlowzero-retention$0.86 in/1M$3.80 out/1M49 tok/s262Kfp812m agoNovita AIzero-retention$0.91 in/1M$3.84 out/1M2 tok/s262Kint412m agoFireworks AIzero-retention$0.95 in/1M$4.00 out/1M51 tok/s262K—12m agoMoonshot AIzero-retention$0.95 in/1M$4.00 out/1M3 tok/s262Kint412m agoTogether AIzero-retention$0.95 in/1M$4.00 out/1M215 tok/s262K—12m agoMoonshot AIhighspeed tier$1.90 in/1M$8.00 out/1M118 tok/s262Kint412m ago
DeepInfra
Novita AI
Cloudflare Workers AI
Nebius AI Studio
Fireworks AIFull specs, hardware verdicts and benchmarks on theKimi K2.7 Code model page →