Skip to main content

​
Overview

glm-4.5-air is the cheap agentic GLM option on Kyma. It is a good fit for bulk automation, repeated agent runs, and cost-sensitive workloads that still need tool use and structured outputs.

​
Specs

FieldValue
Model IDglm-4.5-air
Best forBulk agents, cheap automation, long-context tasks
Context window131K
Max output tokens8K
Input modalitiesText
Output modalitiesText
Tool callingYes
Structured outputsYes
Prompt cachingYes
SpeedFast
Cost bandCheap
Release stageStable

​
Use this when

  • You need an inexpensive model for repeated agent runs.
  • You care about cost more than flagship-level quality.
  • You want tool use and structured outputs in a cheaper lane.

​
Pick something else when

  • You want stronger flagship agent quality: use glm-5.1 or kimi-k2.5.
  • You want the strongest reasoning path: use deepseek-r1.
  • You need multimodal input: use gemma-4-31b or kimi-k2.5.

​
Example

from openai import OpenAI

client = OpenAI(base_url="https://kymaapi.com/v1", api_key="ky-...")

response = client.chat.completions.create(
    model="glm-4.5-air",
    messages=[{"role": "user", "content": "Generate a structured triage summary for these 200 support tickets."}],
    response_format={"type": "json_object"}
)