Home › Models › Z.ai › GLM 5.3

GLM 5.3

GLM 5.3 is Z.ai's precision model for agent plumbing. It is a 400B model with a 1M context window, text and image input, and an intelligence index of 32, the highest GLM score in the fleet. It is the route to use when a response must match a JSON schema or a tool call has to come back well formed. It is not the fleet leader on open-ended visual math; that is Qwen 3.8 Max.

JSON schema adherenceTool callingStructured outputAgent loopsHigh-Frequency Tool Calling & JSON Schema

Input$1.4/1M

Output$4.4/1M

Cached read$0.26/1M

Live telemetry is not published for this route yet.

Parameters400B

Context1M

Intelligence index32

Call it

Other models from Z.ai

GLM 5.1 355B

GLM 5.2 355B