Google has introduced Gemini CLI, a new tool that brings its Gemini AI models directly into the developer’s terminal. Instead of switching between platforms or relying on browser-based assistants, developers can now type natural language commands like “fix this bug” or “write a new function” right where they write code. This approach reflects Google’s push to make AI part of everyday programming, giving developers direct access to advanced AI tools without changing their workflow.
Powered by Gemini 2.5 Pro, the CLI tool can handle large amounts of code, with support for up to one million tokens, enough to understand entire projects. It also works alongside other Google tools, including Code Assist, Search, and creative models like Veo for video and Imagen for image generation. By open-sourcing Gemini CLI under the Apache 2.0 licence, Google hopes to attract developers who want flexibility, transparency, and control over how they use AI.
To support adoption, Google is offering generous free usage: 60 model requests per minute and up to 1,000 per day. This is more than enough for most developers and significantly higher than what competitors currently offer. It also supports enterprise integration via Google Cloud tools like Vertex AI and AI Studio. While tools like OpenAI’s Codex CLI and Anthropic’s Claude Code are also popular, Gemini CLI stands out by being free, deeply integrated with Google services, and open for developer contribution through GitHub.
Despite its capabilities, Google is aware of the limitations of AI coding tools. Surveys from Stack Overflow and studies from universities have shown that AI-generated code can sometimes include bugs or security flaws. Only 43 percent of developers say they fully trust AI code outputs. In response, Google has built in safety features like command approvals and sandboxing to reduce the chance of harmful errors and to help developers stay in control.
By putting its AI tools where developers already work on the command line, Google is trying to build trust and usefulness at the same time. If Gemini CLI can prove itself reliable, secure, and easy to use, it may quickly become a go-to tool for developers looking to combine natural language and coding without leaving their terminal.
