Coding Flow

English

中文

English

中文

Theme

Coding Flow2026-03-16
370
2.3m

Gemini CLI Setup Guide ​

Install and configure the Gemini CLI tool to connect to the Coding Flow platform.

Gemini CLI is Google's official terminal-based AI assistant tool. This guide walks you through the installation and configuration process.

Before you begin, complete the Node.js Environment Installation Guide.

Prerequisites

  • A valid API Key (e.g. sk-******)
  • An accessible service address (Base URL) (usually provided by your deployment/gateway)

Common service Endpoint for This Site

Step 1: Install Gemini CLI ​

Windows
npm install -g @google/gemini-cli
Linux
sudo npm install -g @google/gemini-cli
macOS(npm)
sudo npm install -g @google/gemini-cli
macOS(Homebrew)
brew install gemini-cli

Verify the installation

After the installation is complete, run the following command. If a version number is displayed, the installation was successful.

Bash
gemini --version

Step 2: Create the configuration ​

Set environment variables so Gemini CLI can connect to the Coding Flow platform. You need to configure three variables:

  • GOOGLE_GEMINI_BASE_URL — Service endpoint
  • GEMINI_API_KEY — Your API key
  • GEMINI_MODEL — The model to use

Temporary setup (current session) ​

Windows
$env:GOOGLE_GEMINI_BASE_URL = "https://api.codingflow.ai"
$env:GEMINI_API_KEY = "YOUR_API_KEY"
$env:GEMINI_MODEL = "gemini-3-pro-preview"
Linux
export GOOGLE_GEMINI_BASE_URL="https://api.codingflow.ai"
export GEMINI_API_KEY="YOUR_API_KEY"
export GEMINI_MODEL="gemini-3-pro-preview"
macOS
export GOOGLE_GEMINI_BASE_URL="https://api.codingflow.ai"
export GEMINI_API_KEY="YOUR_API_KEY"
export GEMINI_MODEL="gemini-3-pro-preview"

Permanent setup ​

Windows
[System.Environment]::SetEnvironmentVariable("GOOGLE_GEMINI_BASE_URL", "https://api.codingflow.ai", [System.EnvironmentVariableTarget]::User)
[System.Environment]::SetEnvironmentVariable("GEMINI_API_KEY", "YOUR_API_KEY", [System.EnvironmentVariableTarget]::User)
[System.Environment]::SetEnvironmentVariable("GEMINI_MODEL", "gemini-3-pro-preview", [System.EnvironmentVariableTarget]::User)
Linux
echo 'export GOOGLE_GEMINI_BASE_URL="https://api.codingflow.ai"' >> ~/.bashrc
echo 'export GEMINI_API_KEY="YOUR_API_KEY"' >> ~/.bashrc
echo 'export GEMINI_MODEL="gemini-3-pro-preview"' >> ~/.bashrc
source ~/.bashrc
macOS
echo 'export GOOGLE_GEMINI_BASE_URL="https://api.codingflow.ai"' >> ~/.zshrc
echo 'export GEMINI_API_KEY="YOUR_API_KEY"' >> ~/.zshrc
echo 'export GEMINI_MODEL="gemini-3-pro-preview"' >> ~/.zshrc
source ~/.zshrc

Important

Replace YOUR_API_KEY with the API key you obtained from Coding Flow Platform.

Configuration parameter details

  • GOOGLE_GEMINI_BASE_URL: The API endpoint provided by the model provider, fixed to https://api.codingflow.ai/v1.
  • GEMINI_API_KEY: The authentication key, here represented as YOUR_API_KEY.
  • GEMINI_MODEL: The model name.

Step 3: Start using Gemini CLI ​

Bash
gemini