Polarison

Z.ai · open weights

GLM 5.3 API pricing by provider

GLM 5.3 is available from 36 providers. The cheapest is Wafer at $0.0699 input / $1.40 output per 1M tokens (FP4). Z.ai’s own API charges $1.40 / $4.40, 81% more than the cheapest host. The priciest host, Alibaba, charges 10.7x the cheapest.

Prices from the OpenRouter API, retrieved October 26, 2026. Weights: zai-org/GLM-5.3.

ProviderPrecisionInput / 1MOutput / 1MDocument Q&A / monthContextMax output
Waferfp4$0.0699$1.40$69.961.05M944K
Waferfp4$0.179$1.40$113.601.05M944K
Sail Researchfp8$0.50$2.00$260.001.05M944K
DekaLLM—$0.25$2.95$188.501.05M944K
InferenceNetfp4$0.08$3.60$140.001.05M944K
Novitafp8$0.63$1.98$311.401.05M131K
DeepInfrafp4$0.5625$2.50$300.001.05M131K
Tenstorrentfp8$0.60$2.50$315.001.04M131K
SiliconFlowfp8$0.70$2.20$346.001.05M262K
GMICloudfp8$0.70$2.20$346.001.05M944K
Relace—$0.0703$4.384$159.651.05M131K
Reka—$0.14$4.20$182.001.05M262K
AkashMLfp8$0.19$4.40$208.001.05M131K
Phala—$0.84$2.64$415.201.05M944K
Decartfp4$0.84$2.71$417.301.05M944K
DigitalOcean—$0.91$2.86$449.801.05M128K
Fireworks—$0.98$3.08$484.401.05M944K
PrimeIntellect—$0.98$3.08$484.401.05M131K
Morphfp8$0.73$4.752$434.561.05M944K
Inceptronfp4$1.20$3.50$585.001.05M944K
Alibaba—$1.19$3.74$588.201M131K
CoreWeavefp8$1.20$4.00$600.001.05M944K
Friendli—$1.26$3.96$622.801.05M944K
Mistralnvfp4$1.40$4.40$692.001.05M131K
Baidufp8$1.40$4.40$692.001.05M131K
BaseTenfp4$1.40$4.40$692.001.05M262K
Nebiusfp4$1.40$4.40$692.001.02M922K
Crusoefp4$1.40$4.40$692.001.05M944K
Venice—$1.40$4.40$692.001M131K
Together—$1.40$4.40$692.001.05M944K
Parasailfp8$1.40$4.40$692.001.05M944K
Modal—$1.40$4.40$692.001.05M944K
Fireworks—$1.40$4.40$692.001.05M944K
Cloudflare—$1.40$4.40$692.001.05M944K
AtlasCloudfp8$1.40$4.40$692.001.05M131K
Z.AIcreatorfp8$1.40$4.40$692.001.05M131K
Mistralnvfp4$1.54$4.84$761.201.05M131K
Amazon Bedrock—$1.68$5.28$830.401.05M131K
ModelRunfp4$2.10$6.60$1,0381.05M524K
Fireworks—$2.10$6.60$1,0381.05M944K
BaseTenfp8$2.10$6.60$1,0381.05M262K
Alibaba—$2.80$8.80$1,3841M131K

Sorted by blended price (3 input : 1 output). Document Q&A assumes 50,000 questions a month with 8,000 input and 600 output tokens each. Precision is reported by the provider; “—” means not disclosed.

Frequently asked questions

What is the cheapest GLM 5.3 API provider?
Wafer is the cheapest healthy provider we track at $0.0699 per 1M input tokens and $1.40 per 1M output tokens.
How much does GLM 5.3 cost?
Across 36 providers, input prices range from $0.0699 to $2.80 per 1M tokens and output prices from $1.40 to $8.80.
What is GLM 5.3’s context window?
GLM 5.3 supports up to 1.05M tokens, but some providers serve a smaller context window — check the table before choosing.

Other open-weight models