GLM52.pro

GLM 5.2 + OpenCode

Configure GLM 5.2 as your OpenCode backend for cheap, high-quality AI-assisted coding

Why GLM 5.2 for OpenCode?

53× cheaper than Fable
Run thousands of OpenCode sessions for the price of one Fable session
Coding Plan built-in
GLM 5.2 plans multi-file edits before executing — ideal for OpenCode workflows
Fast responses
~85 tokens/sec means less waiting between file edits

Setup via OpenRouter (Recommended)

1. Get an OpenRouter API key

Sign up at openrouter.ai and add credits. GLM 5.2 output costs $0.28/M tokens.

2. Configure OpenCode
# opencode.json
{
  "model": {
    "provider": "openai",
    "name": "zhipuai/glm-4",
    "baseURL": "https://openrouter.ai/api/v1",
    "apiKey": "sk-or-YOUR_KEY_HERE"
  }
}
3. Start OpenCode
opencode

Setup via Z.ai (Direct)

# opencode.json (Z.ai direct)
{
  "model": {
    "provider": "openai",
    "name": "glm-4-plus",
    "baseURL": "https://open.bigmodel.cn/api/paas/v4",
    "apiKey": "YOUR_ZHIPU_API_KEY"
  }
}

Get your API key at z.ai. Offers slightly lower latency than OpenRouter.

Recommended OpenCode Settings for GLM 5.2

{
  "model": { ... },
  "autoshare": false,
  "keybinds": {
    "leader": "ctrl+x"
  }
}