Google has launched an open-source command-line interface (CLI) called Gemini CLI free of charge to developers.
Google has launched an open-source command-line interface (CLI) called Gemini CLI.
The machine-learning software will bring the power of its Gemini AI models directly to developers’ terminals, and will enable natural-language interaction with Gemini 2.5 Pro from the terminal.
Designed for software engineers and technical users, Gemini CLI supports tasks ranging from code generation, debuggin and file manipulation to general-purpose queries such as content creation or research.
Gemini CLI’s key features include Gemini 2.5 Pro integration, offering up to a 1 million–token context window.
Access is free through personal Google accounts, with usage limits set at 60 requests per minute and 1,000 per day during preview.
Google says Gemini CLI is lightweight and extensible, and will offer support for grounding queries via Google Search and plug-ins using Model Context Protocol (MCP) or bundled extensions.
The software is free and open-source under the Apache 2.0 license, and developers can inspect, modify, and contribute via GitHub.
Gemini CLI expands Google’s AI tools beyond IDEs and web apps, directly into developer environments.
For programmers who rely heavily on the terminal, it provides a flexible, powerful interface for AI-assisted coding – complementing tools like Gemini Code Assist, which Google recently updated to support the same models in Visual Studio Code and JetBrains platforms.
In preview mode, Gemini CLI invites community contributions and security audits via its GitHub portal.
Google plans to enhance its extensibility with more plug-in support and deeper integration into its developer ecosystem.
As AI tools grow in popularity, Gemini CLI signals Google’s intent to embed intelligent agents across all development platforms.
Google introduces open-source AI agent Gemini CLI
