modelbenchmark.io

GLM-5.2 Ponytail

Zhipu · glm-5-2-ponytail

Compare

glm-5.2 carrying a built-in ruleset that compresses generated code, preferring platform features over custom code. The middle tier, and the ruleset as its authors wrote it. Same upstream model and price per token as glm-5.2, with fewer output tokens.

Specification

most-agreed values

Context
1M
Max output
131K
Released
2026-06-13
Knowledge cutoff
—
Retires
—
Open weights
yes
Input
text
Output
text

Price

US dollars per million tokens · most-agreed

Input
$1.7383
Output
$5.1589
Cache read
$0.4486

Available from 1 host

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
greenpt$1.7383$5.1589$0.4486—1M131K—

Price history

append-only observations · a listing writes a row only when its price moves

DateHostField$/M
2026-09-02greenptCache read$0.3135
2026-10-22greenptCache read$0.4486
2026-09-02greenptInput$1.254
2026-10-22greenptInput$1.7383
2026-09-02greenptOutput$5.016
2026-10-22greenptOutput$5.1589

More from Zhipu

most-hosted first