GLM-5.2 API: pricing, base URL and quick setup

Specifications and pricing

Prices synced with platform billing (USD per 1M tokens).

Quick setup

curl https://a2agent.me/v1/chat/completions \
  -H "content-type: application/json" \
  -H "authorization: Bearer sk-your-a2agent-key" \
  -d '{
    "model": "glm-5.2",
    "messages": [{"role": "user", "content": "Reply with OK"}]
  }'

Using it from a coding client

The 1M window is what makes whole-repository prompts practical in Claude Code or Cursor. Pin the exact model id in your config, because glm-5 and glm-5.2 are different models with different windows.

• Run it in Claude Code
• Run it in Cursor

How it compares within its family

GLM-5.2 was the first Z.ai model sold here to reach a 1M context, up from the 200K of GLM-5 and GLM-5.1. It lists at $1.40 per million input tokens and $4.40 output. GLM-5.3 lists at exactly those two numbers, with the same window and the same coding tag, so the catalogue draws no line between them beyond the version number. Treat 5.2 as the pinned, older release: choose it when you have already tested against it and want the behaviour to stop moving, and take 5.3 when you want whatever Z.ai shipped last. Against GLM-5.1 the trade is clearer. Five times the window for roughly twenty percent more per token.

Other models on this platform

• GLM-5200K
• GLM-5.1200K
• GLM-5.31M
• GLM-5.3 Flash1M

Questions