Gemini CLI - API易文档中心

Overview

Gemini CLI is Google’s official command-line tool that allows you to interact directly with Gemini AI models in your terminal. Through API易, you can:

🚀 Quickly call Gemini models in terminal
💻 AI-assisted programming and code generation
🔍 Code review and optimization suggestions
📝 Technical Q&A and documentation generation
🌐 Cross-platform support (Linux, macOS, Windows)

Why Choose API易?Using Gemini CLI through API易 provides more stable network connections, better pricing, and 24/7 technical support.

Quick Start

Prerequisites

Node.js >= 18.0.0
npm or yarn package manager
API易 account and API Key

Step 1: Install Gemini CLI

# Check Node.js version
node --version  # Must be >= 18

# Install Gemini CLI globally
npm install -g @google/gemini-cli

# Verify installation
gemini --version

Step 2: Get API易 API Key

Visit api.apiyi.com to register or login

Create API Key

Go to dashboard “Token Management” page (api.apiyi.com/token) and click “Create New Token”

Copy Key

Copy the generated API Key (format: sk-***) and keep it safe

Step 3: Configure Environment Variables

Important Configuration: GOOGLE_GEMINI_BASE_URL must be set to https://api.apiyi.com without any additional paths (like /v1 or /gemini), otherwise connection will fail.

Zsh (macOS/Linux)
Bash (Linux)
PowerShell (Windows)
CMD (Windows)

# Edit .zshrc file
nano ~/.zshrc

# Add these environment variables
export GOOGLE_GEMINI_BASE_URL="https://api.apiyi.com"
export GEMINI_API_KEY="sk-your-api-key"  # Replace with your API易 key

# Reload configuration
source ~/.zshrc

# Edit .bashrc file
nano ~/.bashrc

# Add these environment variables
export GOOGLE_GEMINI_BASE_URL="https://api.apiyi.com"
export GEMINI_API_KEY="sk-your-api-key"  # Replace with your API易 key

# Reload configuration
source ~/.bashrc

# Set environment variables
$env:GOOGLE_GEMINI_BASE_URL="https://api.apiyi.com"
$env:GEMINI_API_KEY="sk-your-api-key"

# Save permanently (optional)
[System.Environment]::SetEnvironmentVariable('GOOGLE_GEMINI_BASE_URL', 'https://api.apiyi.com', 'User')
[System.Environment]::SetEnvironmentVariable('GEMINI_API_KEY', 'sk-your-key', 'User')

# Temporary environment variables
set GOOGLE_GEMINI_BASE_URL=https://api.apiyi.com
set GEMINI_API_KEY=sk-your-api-key

# Save permanently (requires admin)
setx GOOGLE_GEMINI_BASE_URL "https://api.apiyi.com"
setx GEMINI_API_KEY "sk-your-api-key"

Step 4: Initialize and Test

Launch Gemini CLI

gemini

First-time Authentication

In the interactive interface, type:

/auth

Select: Gemini API Key (AI Studio)

Test Connection

# Simple test
gemini "Hello, test connection"

# Programming-related test
gemini "Explain how React Hooks work"
gemini "Write a Python function to calculate Fibonacci sequence"

Core Features

Code Generation

Function Generation
Project Scaffolding
Unit Tests

gemini "Write a quick sort algorithm in Python with detailed comments"

Example Output:

def quick_sort(arr):
    """
    Quick sort algorithm
    Time complexity: Average O(n log n), Worst O(n²)
    Space complexity: O(log n)
    """
    if len(arr) <= 1:
        return arr

    pivot = arr[len(arr) // 2]
    left = [x for x in arr if x < pivot]
    middle = [x for x in arr if x == pivot]
    right = [x for x in arr if x > pivot]

    return quick_sort(left) + middle + quick_sort(right)

gemini "Create an Express.js REST API project structure with user auth and database config"

gemini "Write Jest unit tests for this function:
function fibonacci(n) {
  if (n <= 1) return n;
  return fibonacci(n - 1) + fibonacci(n - 2);
}"

Code Review

# Review code quality
gemini "Review the following code for performance issues and potential bugs:
[paste your code]
"

# Security audit
gemini "Check this code for security vulnerabilities, especially SQL injection and XSS"

# Best practices
gemini "What can be improved in this React component? Does it follow best practices?"

Technical Q&A

# Concept explanation
gemini "Explain JavaScript closures with practical use cases"

# Error troubleshooting
gemini "Why isn't my Promise being resolved correctly?"

# Performance optimization
gemini "How to optimize React component rendering performance?"

# Architecture design
gemini "Compare microservices vs monolithic architecture"

Documentation Generation

# Generate README
gemini "Generate a professional README.md for my Node.js library with installation, usage, and API docs"

# API documentation
gemini "Generate OpenAPI 3.0 spec documentation for this REST API endpoint"

# Code comments
gemini "Add detailed JSDoc comments to the following code"

Interactive Commands

In Gemini CLI interactive mode, you can use these commands:

Command	Description	Example
`/auth`	Re-authenticate	`/auth`
`/model`	Switch model	`/model gemini-3-pro-preview`
`/clear`	Clear conversation history	`/clear`
`/help`	Show help information	`/help`
`/exit`	Exit CLI	`/exit` or `Ctrl+C`
`/save`	Save conversation to file	`/save conversation.txt`

Model Switching: Use /model command to switch between different Gemini models like gemini-3-pro-preview, gemini-2.5-flash, gemini-2.5-pro, etc.

Supported Models

Through API易, you can use the latest Gemini models:

Gemini 3 Series (Recommended)

Model	Use Case	Features
gemini-3-pro-preview	High-quality code, complex reasoning	🏆 Top performance, #1 on LMArena leaderboard
gemini-3-pro-preview-thinking	Ultra-complex reasoning, algorithm design	🧠 Chain-of-thought output, deep reasoning

Gemini 2.5 Series

Model	Use Case	Features
gemini-2.5-pro	Professional code generation	⚡ High performance, 1M context
gemini-2.5-flash	Fast response, daily development	🚀 Fast speed, low cost
gemini-2.5-flash-lite	Lightweight tasks, batch calls	💰 Ultra-low cost, high-frequency

Recommended Configuration:

Complex programming, architecture design: gemini-3-pro-preview
Daily code generation, Q&A: gemini-2.5-flash
Batch processing, rapid iteration: gemini-2.5-flash-lite

View Complete Model List

View all Gemini models supported by API易, detailed pricing and performance comparison

Advanced Usage

VS Code Integration

Use Gemini CLI extension in VS Code:

{
  "gemini.apiKey": "sk-your-api-key",
  "gemini.baseUrl": "https://api.apiyi.com",
  "gemini.model": "gemini-3-pro-preview",
  "gemini.temperature": 0.7,
  "gemini.maxTokens": 4000
}

GitHub Actions Integration

Automated code review:

name: Gemini Code Review
on:
  pull_request:
    branches: [ main ]

jobs:
  review:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3

      - uses: actions/setup-node@v3
        with:
          node-version: '18'

      - name: Install Gemini CLI
        run: npm install -g @google/gemini-cli

      - name: Run Code Review
        env:
          GEMINI_API_KEY: \${{ secrets.GEMINI_API_KEY }}
          GOOGLE_GEMINI_BASE_URL: https://api.apiyi.com
        run: |
          gemini "Review this Pull Request for code quality, security, and performance:
          $(git diff origin/main...HEAD)"

Batch Scripts

Create automation scripts:

#!/bin/bash

# Batch code review
for file in src/**/*.js; do
  echo "Reviewing $file..."
  gemini "Review code quality of $file" < "$file"
done

# Generate project documentation
gemini "Generate technical documentation outline for the entire project" < README.md

FAQ

Connection failed or authentication error?

Checklist:

Verify environment variables:
```
echo $GOOGLE_GEMINI_BASE_URL
echo $GEMINI_API_KEY
```
Should output:
- GOOGLE_GEMINI_BASE_URL: https://api.apiyi.com (no /v1 or other paths)
- GEMINI_API_KEY: Complete key starting with sk-
Reload environment variables:
```
source ~/.zshrc  # or source ~/.bashrc
```
Restart terminal: Close and reopen terminal completely
Verify API Key: Login to API易 dashboard to confirm key is valid with sufficient balance

How to switch between Gemini models?

Use /model command in interactive mode:

/model gemini-3-pro-preview
/model gemini-3-pro-preview-thinking
/model gemini-2.5-flash

Or specify directly in command:

gemini --model gemini-3-pro-preview "your question"
gemini --model gemini-2.5-flash "quick test"

How to save conversation history?

Method 1: Use /save command

/save conversation-2025-01-01.txt

Method 2: Redirect output

gemini "your question" > output.txt
gemini "your question" | tee output.txt  # Display and save

Method 3: Use session management

gemini --session my-project "continue previous discussion"

Node.js version doesn't meet requirements?

Recommended: Use nvm to manage Node.js versions:

# Install nvm
curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.39.0/install.sh | bash

# Install Node.js 18+
nvm install 18
nvm use 18
nvm alias default 18

# Verify version
node --version

Why is response slow?

Possible causes and solutions:

Network issues: API易 provides optimized domestic nodes, usually fast
Model selection: Use gemini-2.5-flash or gemini-2.5-flash-lite for fastest response
Token limit: Reduce complexity of single request
Concurrent requests: Avoid sending too many requests simultaneously

Test connection speed:

time gemini --model gemini-2.5-flash "Hello"

How to use on Windows?

Recommended: Use PowerShell:

Install Node.js (download installer from official website)
Run PowerShell as administrator

Set environment variables:

[System.Environment]::SetEnvironmentVariable('GOOGLE_GEMINI_BASE_URL', 'https://api.apiyi.com', 'User')
[System.Environment]::SetEnvironmentVariable('GEMINI_API_KEY', 'sk-your-key', 'User')

Restart PowerShell
Install and use Gemini CLI

Or use WSL (Windows Subsystem for Linux) for better experience.

Best Practices

Prompt Optimization

Be Specific

❌ “Optimize this code”✅ “Optimize this code for performance, focusing on loop efficiency and memory usage”

Provide Context

❌ “What’s wrong with this function?”✅ “This is a user login function encountering async errors, help me find the issue”

Step-by-Step

❌ “Help me complete the entire project”✅ “Step 1: Design database models; Step 2: Create API routes; Step 3: …”

Request Examples

❌ “Explain closures”✅ “Explain JavaScript closures with 3 practical use cases and code examples”

Workflow Recommendations

Define Problem

Clearly describe the problem to solve or feature to implement

Get Solution

Use Gemini CLI to generate initial solution or code

Review & Optimize

Ask AI to review its own code and find potential issues

Iterate

Gradually optimize based on feedback until requirements are met

Add Documentation

Generate necessary comments and documentation

Pricing

Cost of using Gemini models through API易 depends on your chosen model and usage.

View Detailed Pricing

View detailed pricing and cost-effectiveness comparison for all Gemini models

API易 offers recharge bonuses: the more you recharge, the higher the bonus (10%-22%). First-time recharge gets extra bonus. View recharge promotion details.

Get Help

Enterprise WeChat

Scan QR code or Click to contact supportTechnical consultation, usage guidance

Email Inquiry

Customer Service: support@apiyi.comBusiness: business@apiyi.com

Quick Start: Follow the “Quick Start” section above to complete setup and start using Gemini CLI in 5 minutes!

Overview

AI Agent

Conversational AI

Programming

Engineering

Translation

Community Contributions

​Overview

​Quick Start

​Prerequisites

​Step 1: Install Gemini CLI

​Step 2: Get API易 API Key

​Step 3: Configure Environment Variables

​Step 4: Initialize and Test

​Core Features

​Code Generation

​Code Review

​Technical Q&A

​Documentation Generation

​Interactive Commands

​Supported Models

​Gemini 3 Series (Recommended)

​Gemini 2.5 Series

View Complete Model List

​Advanced Usage

​VS Code Integration

​GitHub Actions Integration

​Batch Scripts

​FAQ

​Best Practices

​Prompt Optimization

Be Specific

Provide Context

Step-by-Step

Request Examples

​Workflow Recommendations

​Pricing

View Detailed Pricing

​Related Resources

​Get Help

Enterprise WeChat

Email Inquiry

Overview

Quick Start

Prerequisites

Step 1: Install Gemini CLI

Step 2: Get API易 API Key

Step 3: Configure Environment Variables

Step 4: Initialize and Test

Core Features

Code Generation

Code Review

Technical Q&A

Documentation Generation

Interactive Commands

Supported Models

Gemini 3 Series (Recommended)

Gemini 2.5 Series

Advanced Usage

VS Code Integration

GitHub Actions Integration

Batch Scripts

FAQ

Best Practices

Prompt Optimization

Workflow Recommendations

Pricing

Related Resources

Get Help