Google Releases Gemini CLI: Free and Open-Source AI Agent for Developers
Big news coming from Google’s AI team integrating Gemini AI right into Command Line Interface (CLI) for free! Today, Google unveiled a free and open source Gemini CLI AI agent for developers which can essentially write code for you in real-time. This comes handy for creating or developing any type of app that requires CLI.
The best part is, once you sign in with your personal Google account. This will grant you up to 60 model requests per minute and 1,000 model requests per day using Gemini 2.5 Pro. Generous amount of requests for free.
The command line interface has long been the sacred sanctuary of developers—a place where efficiency meets elegance, and where the most complex problems are solved with simple keystrokes. Today, Google has transformed this familiar territory into something extraordinary with the launch of Gemini CLI, an open-source AI agent that brings the full power of Google’s advanced Gemini models directly into your terminal.
This is reimagining of how we interact with artificial intelligence in our daily coding workflows. Google has created what may be the most accessible and powerful AI coding assistant available today.
What is Gemini CLI?
Gemini CLI is a command-line AI workflow tool that connects to your tools, understands your code and accelerates your workflows.
With the Gemini CLI you can:
- Query and edit large codebases in and beyond Gemini’s 1M token context window.
- Generate new apps from PDFs or sketches, using Gemini’s multimodal capabilities.
- Automate operational tasks, like querying pull requests or handling complex rebases.
- Use tools and MCP servers to connect new capabilities, including media generation with Imagen, Veo or Lyria
- Ground your queries with the Google Search tool, built in to Gemini.
The Terminal as Your AI Command Center
Gemini CLI is a paradigm shift in AI accessibility for developers. Unlike web-based interfaces or IDE-specific plugins, this tool lives where developers actually work—in the terminal. The beauty of this approach lies in its universality. Whether you’re working on a MacBook, a Linux server, or a Windows machine, your terminal becomes an instant gateway to advanced AI capabilities.
The tool excels at traditional coding tasks, but its versatility extends far beyond code generation. Developers can leverage Gemini CLI for content creation, complex problem-solving, research tasks, and even project management—all without leaving their command line environment.
Take a look at how Gemini CLI works in the video below:
Unprecedented Usage Limits
One of the most compelling aspects of Gemini CLI is Google’s approach to usage limits. While many AI tools restrict users with tight rate limits that interrupt workflow, Google has taken a different path. With a simple personal Google account login, developers gain access to:
- 60 model requests per minute – enough to maintain fluid conversation and rapid iteration
- 1,000 requests per day – supporting even the most AI-intensive development sessions
- Access to Gemini 2.5 Pro – including its massive 1 million token context window
- Completely free of charge – removing financial barriers to AI-assisted development
These limits aren’t just generous; they’re practically designed to ensure that rate limiting never becomes a roadblock during active development work. For professional developers requiring even more capacity, the tool integrates seamlessly with Google AI Studio and Vertex AI for usage-based billing.
Technical Capabilities
Advanced Model Integration
Gemini CLI provides access to some of Google’s most sophisticated AI models, including Gemini 2.5 Pro with its industry-leading context window. This massive context capacity means developers can work with entire codebases, lengthy documentation, or complex project specifications without losing context or requiring constant re-prompting.
Built-in Tools and Extensions
The platform comes equipped with powerful built-in capabilities that extend its functionality:
Google Search Integration: Ground your prompts with real-time web data, allowing the AI to fetch current information and provide contextually relevant responses based on the latest developments in your field.
Model Context Protocol (MCP) Support: Built on emerging standards, Gemini CLI supports extensible functionality through MCP, ensuring compatibility with a growing ecosystem of AI tools and services.
Custom Prompt Engineering: Developers can customize instructions and prompts to align with specific workflows, coding standards, or project requirements, making the AI truly personalized to individual or team needs.
Script Integration: The tool can be invoked non-interactively, enabling automation and integration with existing CI/CD pipelines, build scripts, and development workflows.
Open Source
Apache 2.0 License Advantage
By releasing Gemini CLI under the Apache 2.0 license, Google has made a significant statement about transparency and community collaboration. Developers can inspect the entire codebase, understand exactly how the tool works, and verify its security implications—crucial factors for enterprise adoption and security-conscious organizations.
Community-Driven Development
The open-source nature of Gemini CLI opens the door for global community contributions. This means the tool will evolve based on real developer needs, with bug reports, feature requests, and code improvements coming directly from the people who use it daily. The GitHub repository serves as a central hub for collaboration, ensuring that the tool remains responsive to the evolving needs of the development community.
Extensibility and Customization
The tool’s architecture supports extensive customization through system prompts (via GEMINI.md files) and both personal and team configuration settings. This recognition that every developer’s terminal environment is unique ensures that Gemini CLI can adapt to diverse workflows and preferences.
Seamless IDE Experience
While Gemini CLI excels in terminal environments, Google recognizes that sometimes an IDE is the right tool for the job. The integration with Gemini Code Assist brings the same powerful AI capabilities to VS Code, creating a unified experience across different development environments.
How to get Gemini in CLI?
Being opens source, Gemini CLI code is readily available on Github repository. All you need is Node.js version 18 or higher installed. You can then start with Google CLI with the following lines:
- Prerequisites: Ensure you have Node.js version 18 or higher installed.
- Run the CLI: Execute the following command in your terminal:
npx https://github.com/google-gemini/gemini-cli
- Or install it with:
npm install -g @google/gemini-cli
- Pick a color theme
- Authenticate: When prompted, sign in with your personal Google account. This will grant you up to 60 model requests per minute and 1,000 model requests per day using Gemini 2.5 Pro.
