Gemini CLI Installation & Setup Guide: Free AI Coding Tool Configuration Tutorial

Introduction: Why Choose Gemini CLI

As developers around you embrace AI-assisted coding tools, are you still sticking to "old-school programming"? Among the mainstream AI coding tools on the market, Claude Code is powerful but comes with account suspension risks and a hefty price tag. Google's Gemini CLI offers a free and powerful alternative — built on the Gemini 2.5 Pro model, it provides 1,000 free requests per day, making it essentially a zero-cost AI programming assistant for individual developers.

About Gemini 2.5 Pro: This is Google DeepMind's flagship multimodal large language model released in 2025, excelling in code generation, mathematical reasoning, and long-context understanding. With a context window of up to 1 million tokens, it can process the equivalent of a thick book's worth of codebase in one go — particularly valuable for refactoring and analyzing large projects. In mainstream coding benchmark evaluations (such as HumanEval and SWE-bench), Gemini 2.5 Pro ranks in the same tier as Claude 3.5 Sonnet and GPT-4o, but Google's free API quota strategy significantly lowers the barrier to entry.

The current AI coding tool market has formed a multi-player landscape: Anthropic's Claude Code focuses on deep code understanding and refactoring, OpenAI's Codex series is deeply integrated into GitHub Copilot, Microsoft has built a massive user base through its VS Code plugin ecosystem, and Cursor has risen rapidly as an AI-native IDE. Gemini CLI's differentiation strategy lies in: entering via a command-line interface, not being tied to any specific editor, and offering generous free quotas — primarily targeting developers who prefer terminal operations and newcomers looking to try AI coding at low cost.

This article will walk you through the complete installation and configuration of Gemini CLI, followed by a practical demo showcasing its programming capabilities.

Installing Node.js: Prerequisites for Running Gemini CLI

Gemini CLI runs on Node.js, so you first need to ensure your system has Node.js installed. If your terminal shows a "command not found" error for npm, you'll need to install Node.js first.

Introduction to Node.js and npm: Node.js is a JavaScript runtime environment built on Chrome's V8 engine. Since its release in 2009, it has become core infrastructure for both frontend and backend development. npm (Node Package Manager) is the package management tool bundled with Node.js, currently hosting over 2 million open-source packages as the world's largest software registry. Simply put, npm is to Node.js what the App Store is to iPhone — it lets developers install and manage various tools and dependency libraries with a single command.

Node.js Download and Installation Steps

1. Go to the Node.js official website (https://nodejs.org) and download the installer for your operating system
2. Mac users can directly download the .pkg file and double-click to run it
3. Follow the installation wizard: click "Continue" → "Agree" → enter your password → complete the installation

After installation, type node -v and npm -v in your terminal to verify the installation was successful. Seeing version number output means your environment is ready.

Installing Gemini CLI: One Command Is All You Need

With the environment ready, you can now install Gemini CLI. Google provides a one-command installation method via npm.

Gemini CLI Installation Command

Run the following command in your terminal:

npm install -g @google/gemini-cli

If you encounter permission errors, Mac/Linux users need to prepend sudo:

sudo npm install -g @google/gemini-cli

Command Breakdown: The -g flag in npm install -g stands for global installation, meaning the tool will be installed in a system-level directory and can be called from any path, not just the current project directory. This is why after installing once, you can type the gemini command from any folder on your computer to launch it.

Enter your system password and wait for the installation to complete. When running the install command again produces no output, the installation was successful.

Creating a Working Directory

After installation, it's recommended to create a dedicated working directory for testing:

mkdir gemini-test
cd gemini-test

Then type gemini in that directory to launch the CLI tool. When the interactive interface appears, the installation is confirmed successful.

Gemini CLI Account Authentication and Login Configuration

When launching Gemini CLI for the first time, you'll need to complete the Google account authentication process.

Google Account Authentication Steps

1. Select login method: The system will prompt you to choose a verification method — just select the first option (browser login)
2. Browser authorization: The system will automatically open your browser; select a Google account that has Gemini service enabled
3. Confirm authorization: After completing authorization in the browser, the terminal will display "You were successfully signed into Google"
4. Restart CLI: Press the R key to restart Gemini CLI

After restarting, you'll see the terminal showing a connection to the Gemini model, and you're ready to start using it. You may need to unlock access permissions on first use — just follow the prompts.

Hands-On Demo: Auto-Generating a Lottery Program with Gemini CLI

With configuration complete, let's experience Gemini CLI's AI code generation capabilities through a practical example.

Issuing a Programming Task in Natural Language

Type your requirement directly in Gemini CLI:

Write me a random company annual party lottery program in HTML code, the kind that can run directly

The AI Auto-Coding Process: The Magic of Agent Mode

After receiving the instruction, Gemini CLI enters a "thinking" state for analysis and planning, then automatically begins writing code. Behind the scenes, it's running in AI Agent (intelligent agent) mode: it can not only generate code snippets but also proactively plan task steps, invoke file system operations, execute Shell commands, and request user authorization at critical checkpoints. This "plan → execute → feedback" loop mechanism is what fundamentally differentiates it from ordinary code completion tools (like GitHub Copilot) — the latter only passively responds to context at the cursor position, while Agent-mode tools proactively drive the entire development task to completion.

During this process, it will ask multiple times whether you allow certain operations (such as creating files, writing content, etc.) — you just need to click "Allow."

Throughout the entire process, developer interaction is simplified to the extreme — you only need two keyboard actions: "Allow" and "Execute" — without writing a single line of code yourself.

The Generated Lottery Program Results

Gemini CLI automatically generated a complete index.html file, including:

• Start Lottery button: Names scroll when clicked
• Stop button: Randomly stops on a name when clicked
• Winner notification: A "Congratulations!" animation effect pops up

Simply open this HTML file in a browser to run it — no additional configuration needed. From issuing the requirement to receiving a working product, the entire process takes less than two minutes.

What Gemini CLI Means for Developers

A Qualitative Shift in Development Efficiency

AI coding tools like Gemini CLI bring more than just efficiency improvements — they represent a fundamental shift in development paradigms. The developer's role is transforming from "code writer" to "requirement describer and code reviewer." This shift is known in software engineering as Intent-Driven Programming — developers only need to clearly express "what they want" rather than "how to implement it," while AI fills in the technical details in between.

Significantly Lowering the Programming Barrier

More noteworthy is that these tools are lowering the barrier to entry for programming. Product managers and even operations staff can complete simple development tasks with Gemini CLI, as long as they have a basic understanding of code. Specifically:

• Non-professional developers can quickly implement simple functional requirements
• Professional developers can focus their energy on architecture design and complex logic
• Team collaboration efficiency will see significant improvements

Gemini CLI Core Advantages at a Glance

Feature	Description
Free quota	1,000 requests per day, more than enough for personal use
Model capability	Based on Gemini 2.5 Pro, high-quality code generation
Ease of use	Command-line interaction, just describe requirements in natural language
Stability	Lower account suspension risk compared to some other tools
Agent capability	Proactively plans and executes multi-step tasks, not just code completion

Conclusion

If you're still "old-school programming"