MOUNTAIN VIEW, CA – June 25, 2025 – Google has officially launched Gemini CLI, an open-source AI agent designed to seamlessly integrate the power of Gemini directly into developers’ command-line interfaces. This innovative tool aims to significantly boost productivity and streamline operations for a diverse range of users, from hobbyists to large development teams.
Gemini CLI is a versatile, local utility that extends far beyond traditional coding assistance. Its capabilities encompass content generation, problem-solving, in-depth research, and efficient task management. A major advantage for individual developers is the unparalleled free usage limits: by simply logging in with a personal Google account, users gain access to a complimentary Gemini Code Assist license. This license includes Gemini 2.5 Pro, boasting a substantial 1 million token context window, and provides an industry-leading allowance of 60 model requests per minute and 1,000 requests per day. For professional developers requiring higher usage or specific models, options include utilizing a Google AI Studio or Vertex AI key, or upgrading to a Gemini Code Assist Standard or Enterprise license.
The core strength of Gemini CLI lies in its integrated tools, which empower users to:
- Ground prompts with Google Search: Access real-time external context to enhance prompt relevance.
- Extend capabilities: Utilize the Model Context Protocol (MCP) or bundled extensions for expanded functionality.
- Customize prompts: Tailor prompts to specific project requirements.
- Automate tasks: Invoke Gemini CLI non-interactively within scripts for seamless automation.
Furthermore, Gemini CLI shares its underlying technology with Google’s acclaimed AI coding assistant, Gemini Code Assist. This integration ensures that all developers, regardless of their Code Assist plan, can leverage prompt-driven, AI-first coding within both VS Code and Gemini CLI. The agent mode of Gemini Code Assist, available at no additional cost, offers a multi-step, collaborative reasoning agent capable of building plans, recovering from failed implementations, and recommending solutions for intricate coding challenges.
Released under the Apache 2.0 license, Gemini CLI embraces an open-source model, actively encouraging community contributions. Developers are invited to inspect the code, report bugs, suggest features, and contribute to security and code enhancements. This open and extensible architecture, built upon emerging standards like MCP and system prompts, allows for extensive customization to suit individual preferences and team configurations.
Getting started with Gemini CLI is straightforward, requiring only an email address for practically unlimited use directly from the terminal. This release marks a pivotal moment in making advanced AI capabilities readily accessible to developers, promising to transform the command-line experience.
For more details, you can refer to the official announcement on the Google Blog.
