Provider guide
Where to run GLM 5.2
32 live listings tracked — output prices vary 5.3× between the cheapest and most expensive host, so the provider choice matters as much as the model choice.
Rekazero-retentionCHEAPEST$0.46 in/1M$0.91 out/1M54 tok/s1M—2 hours agoStreamLake$0.56 in/1M$1.76 out/1M52 tok/s1Mfp82 hours agoDeepInfrazero-retention through OpenRouterDirect and through OpenRouter$0.75 in/1M$0.56 in/1M$2.40 out/1Mdirect$1.80 out/1Mthrough OpenRouter53 tok/sthrough OpenRouter1Mfp42 hours agoNovita AIzero-retention through OpenRouterDirect and through OpenRouter$1.40 in/1M$0.65 in/1M$4.40 out/1Mdirect$2.04 out/1Mthrough OpenRouter70 tok/sthrough OpenRouter1Mfp82 hours agoSiliconFlowzero-retention$0.70 in/1M$2.20 out/1M71 tok/s1Mfp82 hours agoDigitalOcean Gradientzero-retention$0.70 in/1M$2.20 out/1M85 tok/s262K—2 hours agoDecartzero-retention$0.39 in/1M$2.40 out/1M59 tok/s1Mmxfp42 hours agoCoreWeavezero-retention$0.76 in/1M$2.42 out/1M8 tok/s1Mfp42 hours agoMorphzero-retention$0.46 in/1M$2.84 out/1M76 tok/s1Mfp82 hours agoAtlasCloud$0.94 in/1M$2.95 out/1M61 tok/s1Mfp82 hours agoPhalazero-retention$1.26 in/1M$3.00 out/1M122 tok/s1Mfp82 hours agoAlibaba Cloud$0.97 in/1M$3.04 out/1M58 tok/s1Mfp82 hours agoOpenRouter$0.41 in/1M$3.99 out/1M—1M—2 hours agoWaferzero-retention$0.41 in/1M$3.99 out/1M85 tok/s1M—2 hours agoRelacezero-retention$0.13 in/1M$4.00 out/1M80 tok/s1M—2 hours agoInceptronzero-retention$1.39 in/1M$4.39 out/1M59 tok/s1Mfp42 hours agoTogether AIzero-retention$1.40 in/1M$4.40 out/1M67 tok/s1M—2 hours agoParasailzero-retention$1.40 in/1M$4.40 out/1M60 tok/s262Kfp42 hours agoMistral AI$1.40 in/1M$4.40 out/1M93 tok/s1Mnvfp42 hours agoCloudflare Workers AI$1.18 in/1M$4.40 out/1M23 tok/s262K—2 hours agoGMICloud$1.40 in/1M$4.40 out/1M15 tok/s1Mfp82 hours agoBaidu$1.40 in/1M$4.40 out/1M57 tok/s1Mfp820 hours agoFireworks AIzero-retention$1.40 in/1M$4.40 out/1M75 tok/s1M—3 days agoFriendli$1.40 in/1M$4.40 out/1M83 tok/s1M—2 hours agoZ.AIzero-retention$1.40 in/1M$4.40 out/1M68 tok/s1Mfp82 hours agoBasetenzero-retention$1.40 in/1M$4.40 out/1M49 tok/s1Mfp82 hours agoVenice AIzero-retention$1.40 in/1M$4.40 out/1M91 tok/s1Mfp82 hours agoMistral AIzero-retention$1.54 in/1M$4.84 out/1M89 tok/s1M—2 hours agoBaidufast tier$1.40 in/1M$4.90 out/1M81 tok/s1Mfp432 hours agoBasetenfast tier$2.10 in/1M$6.60 out/1M109 tok/s1Mfp82 hours agoAlibaba Cloudfast tier$2.31 in/1M$7.26 out/1M48 tok/s1Mfp82 hours agoDecartfast tier$2.25 in/1M$8.00 out/1M162 tok/s1Mfp42 hours ago
DeepInfra
Novita AI
DigitalOcean Gradient
Morph
OpenRouter
Together AI
Mistral AI
Cloudflare Workers AI
Baidu
Fireworks AIFull specs, hardware verdicts and benchmarks on the GLM 5.2 model page →