modelbenchmark.io

GLM-5.3

Zhipu · glm-5-3-fast

Compare

Flagship GLM model for long-horizon coding, agents, and complex project delivery

Specification

most-agreed values

Context
1M
Max output
131K
Released
2026-08-14
Knowledge cutoff
—
Retires
—
Open weights
yes
Input
text
Output
text

Price

US dollars per million tokens · most-agreed

Input
$2.80
Output
$8.80
Cache read
$0.56
Cache write
$1.68

Available from 9 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
coralbricks · coralbricks$1.12$4.40free$1.681M1M—
baseten$2.10$6.60$0.21—1M262K—
baseten · zai-org$2.10$6.60$0.21—1M262K—
fireworks_ai$2.10$6.60$0.39—1M1M—
fireworks-ai · accounts$2.10$6.60$0.39—1M262K—
vercel · zai$2.10$6.60$0.21—1M262K—
wally$2.10$6.60$0.39—1M131K—
llmgateway$2.80$8.80$0.56—1M131K—
llmgateway-providers · scx-ai-gp$2.80$8.80$0.56—1M128K—

Other listings of this model

this page is one of them

The corpus files GLM-5.3 under several keys. This page is the glm-5-3-fast listing. The full record — every host, every price and every benchmark score — is on the main GLM-5.3 page.

More from Zhipu

most-hosted first