Z.AI GLM 5 inference overview

Price per 1M tokens

$1.00 (input)

$3.20 (output)

Parameters

40B (active)

744B (total)

Context Window

203K

Release Date

Feb 2026

Z.AI GLM 5 inference details

GLM-5 is a Mixture-of-Experts (MoE) model for long-horizon agentic tasks. It delivers significant improvement compared to GLM-4.7 across a wide range of academic benchmarks with strong performance across reasoning, coding, and agentic tasks.

Created by:

Z.AI

License:

mit

Model card:

GLM-5-FP8

Copy

Z.AI GLM 5 resources

Course

AI engineering course: Agents

Guide

W&B Inference powered by CoreWeave

Whitepaper

A primer on building successful AI agents