Gemini CLI is a command-line interface tool that can be made compatible with Z.AI’s GLM models by using a customized fork.

​
Step 1: Obtaining the Custom Version

​
1. Cloning the Custom Repository

Since the official Gemini CLI repository only supports Google’s Gemini models, we need to use a customized branch that supports OpenRouter compatibility: 
git clone https://github.com/heartyguy/gemini-cli  
cd gemini-cli

​
2. Switching to the Compatible Branch

git checkout feature/openrouter-support

​
Step 2: Environment Configuration

​
1. Setting Environment Variables

Description Configure the API base URL: 
export OPENROUTER_BASE_URL="https://api.z.ai/api/coding/paas/v4"
Configure the API Key: 
export OPENROUTER_API_KEY="your_api_key"

​
Step 3: Installation and Launch

​
1. System Requirements

Ensure your Node.js version is >= 18.

​
2. Installing Dependencies

npm install

​
3. Launch Process

Description After launching, complete the following steps:
  1. Select a background color theme.
  2. User login (recommended to use a Google account for authentication, which will redirect to a webpage for verification).

​
Step 4: Usage Results

Description Once configured, you can use Z.AI’s GLM models in the command line for conversations and code generation.