Provider guide
Where to run GLM 5
17 live offers tracked — output prices vary 1.7× between the cheapest and most expensive host, so the provider choice matters as much as the model choice.
FASTEST MEASURED
90 tok/s
measured throughput · $1.00 input / $3.20 output per million tokens
GMICloud$0.60 in/1M$1.92 out/1M49 tok/s203Kfp812m agoStreamLake$0.60 in/1M$1.92 out/1M39 tok/s198Kfp812m agoDeepInfra$0.60 in/1M$2.08 out/1M—203Kfp46h agoBaidu$0.70 in/1M$2.24 out/1M34 tok/s203Kfp812m agoChutes$0.95 in/1M$2.55 out/1M47 tok/s203Kfp86d agoOpenRouterCHEAPEST$0.95 in/1M$2.55 out/1M—205K—14m agoAtlasCloud$0.95 in/1M$3.15 out/1M37 tok/s203Kfp812m agoNovita AI$1.00 in/1M$3.20 out/1M—203K—6h agoDeepInfrazero-retention$0.60 in/1M$2.08 out/1M17 tok/s203Kfp412m agoDigitalOcean Gradientzero-retention$0.75 in/1M$2.40 out/1M7 tok/s64K—12m agoSiliconFlowzero-retention$0.95 in/1M$2.55 out/1M44 tok/s205Kfp812m agoNovita AIzero-retention$1.00 in/1M$3.20 out/1M36 tok/s203Kfp812m agoVenice AIzero-retention$1.00 in/1M$3.20 out/1M14 tok/s198Kfp812m agoParasailzero-retention$1.00 in/1M$3.20 out/1M38 tok/s203Kfp812m agoZ.AIzero-retention$1.00 in/1M$3.20 out/1M32 tok/s203Kfp812m agoAmazon Bedrockzero-retention$1.00 in/1M$3.20 out/1M90 tok/s203K—12m agoPhalazero-retention$1.20 in/1M$3.50 out/1M12 tok/s203K—12m ago
DeepInfra
Baidu
Novita AI
DigitalOcean GradientFull specs, hardware verdicts and benchmarks on theGLM 5 model page →