Z AI2API

How to Set Up and Use Z AI2API

Z AI2API is a proxy server that converts Z.ai Chat (powered by Zhipu AI's GLM models) into an OpenAI-compatible and Anthropic-compatible API format, allowing any tool that speaks the OpenAI API to use Z.ai's GLM-4.5, GLM-4.6, and related models without needing a separate API token. It supports anonymous visitor mode that auto-obtains a random session token, intelligently handles chain-of-thought reasoning output in multiple formats, and enables image uploads for vision models when a real account token is supplied. Developers use it as a local bridge to access Z.ai models from OpenAI-compatible clients and MCP tools.

git clone https://github.com/hmjz100/Z.ai2api.git
cd Z.ai2api

pip install -r requirements.txt

BASE=https://chat.z.ai
PORT=8080
MODEL=GLM-4.5
ANONYMOUS_MODE=true
THINK_TAGS_MODE=reasoning
DEBUG_MODE=false
# Optional: TOKEN=your_z_ai_account_token

python app.py

{
  "mcpServers": {
    "z-ai2api": {
      "command": "python",
      "args": ["/absolute/path/to/Z.ai2api/app.py"],
      "env": {
        "PORT": "8080",
        "ANONYMOUS_MODE": "true",
        "THINK_TAGS_MODE": "reasoning"
      }
    }
  }
}

- "Send a chat completion request to http://localhost:8080/v1/chat/completions using model GLM-4.5"
- "List available models via GET http://localhost:8080/api/models"
- "Upload an image in the messages array to use GLM vision capabilities (requires TOKEN in .env)"