API pricing

Z.ai (Zhipu) API pricing

Z.ai (Zhipu) API prices run from $0.06 per million input tokens (GLM-4.7-Flash) to $2.80 (GLM 5.3 Prime). Its highest-ranked model, GLM-5.3, costs $1.40 input and $4.40 output per million tokens.

Last verified

Official price list: docs.z.ai

API prices per million tokens
Cached input
GLM-4.7-Flash (open weights)$0.06$0.40$0.01$0.15200K38.8
GLM-4.7-FlashX (open weights)$0.07$0.40$0.01$0.15200K—
GLM-5.3-Flash (open weights)$0.15$0.50$0.03$0.241M51.8
GLM-4.5-Air (open weights)$0.20$1.10$0.03$0.43131K38.9
GLM-4.6V (open weights)$0.30$0.90—$0.45128K41.3
GLM-5.3-FlashX (open weights)$0.37$1.25$0.075$0.591M—
GLM-4.5V (open weights)$0.60$1.80—$0.9064K39.8
GLM-4.5 (open weights)$0.60$2.20$0.11$1131K42.0
GLM-4.6 (open weights)$0.60$2.20$0.11$1205K41.4
GLM-4.7 (open weights)$0.60$2.20$0.11$1205K42.0
GLM-5 (open weights)$1$3.20$0.20$1.55205K46.1
GLM-5-Turbo$1.20$4$0.24$1.90200K—
GLM-5V-Turbo$1.20$4$0.24$1.90200K43.8
GLM-5.1 (open weights)$1.40$4.40$0.26$2.15200K47.8
GLM-5.2 (open weights)$1.40$4.40$0.26$2.151M51.1
GLM-5.3 (open weights)$1.40$4.40$0.26$2.151M54.8
GLM 5.3 Prime$2.80$8.80$0.56$4.301M—

Prices on other platforms

The same models are often sold through clouds and resellers at different rates.

ModelRouteInputOutputChecked
GLM-5.3bedrock$1.68$5.282026-10-10
deepinfra$0.90$42026-10-10
fireworks$1.40$4.402026-10-10
mistral$1.40$4.402026-10-10
openrouter$0.039$4.802026-10-10
together$1.40$4.402026-10-10
zai$1.40$4.402026-10-10
GLM-5.3-Flashdeepinfra$0.15$0.502026-10-10
fireworks$0.15$0.502026-10-10
openrouter$0.15$0.502026-10-10
together$0.15$0.502026-10-10
zai$0.15$0.502026-10-10
GLM-5.2alibaba$1.40$4.402026-10-10
deepinfra$0.75$2.402026-10-10
mistral$1.40$4.402026-10-10
openrouter$0.06$72026-10-10
together$1.40$4.402026-10-10
vertex$1.40$4.402026-10-10
zai$1.40$4.402026-10-10
GLM-5.1deepinfra$1.05$3.502026-10-10
openrouter$0.97$3.042026-10-10
together$1.40$4.402026-10-10
zai$1.40$4.402026-10-10
GLM-5bedrock$1$3.202026-10-10
deepinfra$0.60$2.082026-10-10
openrouter$0.60$1.922026-10-10
together$1$3.202026-10-10
vertex$1$3.202026-10-10
zai$1$3.202026-10-10
GLM-5V-Turboopenrouter$1.20$42026-10-10
zai$1.20$42026-10-10
GLM-4.5openrouter$0.60$2.202026-10-10
zai$0.60$2.202026-10-10
GLM-4.7bedrock$0.60$2.202026-10-10
deepinfra$0.40$1.752026-10-10
openrouter$0.60$2.202026-10-10
vertex$0.60$2.202026-10-10
zai$0.60$2.202026-10-10
GLM-4.6deepinfra$0.50$22026-10-10
openrouter$0.50$22026-10-10
zai$0.60$2.202026-10-10
GLM-4.6Vopenrouter$0.30$0.902026-10-10
zai$0.30$0.902026-10-10
GLM-4.5Vopenrouter$0.60$1.802026-10-10
zai$0.60$1.802026-10-10
GLM-4.5-Airopenrouter$0.13$0.852026-10-10
zai$0.20$1.102026-10-10
GLM-4.7-Flashbedrock$0.07$0.402026-10-10
deepinfra$0.06$0.402026-10-10
openrouter$0.0605$0.402026-10-10
GLM-4.7-FlashXzai$0.07$0.402026-10-10
GLM-5.3-FlashXopenrouter$0.37$1.252026-10-10
zai$0.37$1.252026-10-10
GLM 5.3 Primeopenrouter$2.80$8.802026-10-10
GLM-5-Turboopenrouter$1.20$42026-10-10
zai$1.20$42026-10-10

Frequently asked questions

How much does the Z.ai (Zhipu) API cost?

Z.ai (Zhipu) API prices run from $0.06 per million input tokens (GLM-4.7-Flash) to $2.80 (GLM 5.3 Prime). Its highest-ranked model, GLM-5.3, costs $1.40 input and $4.40 output per million tokens.

Does Z.ai (Zhipu) discount cached input?

Yes. Cached input is billed at a lower rate on 15 of the 17 priced models; the cached rate is listed next to each model.