Google has unveiled an AI agent designed to assist in programming. It comprehends prompts in natural language when used with a terminal.
Gemini CLI is our new #OpenSource AI agent that brings the power of Gemini directly into your terminal!
Access Gemini 2.5 Pro with 1M token context window, 60 requests/min, and 1,000 requests/day—at no cost with a free Gemini Code Assist license → https://t.co/Fp0wdxfbEW pic.twitter.com/Ul6hjn1hKQ
— Google Cloud Tech (@GoogleCloudTech) June 25, 2025
Gemini CLI aims to reduce the complexity of traditional programming interfaces, allowing developers to use natural language when interacting with AI.
“With Gemini CLI, you can naturally communicate with your computer, solve problems, integrate complex workflows, and do much more than before,” noted Taylor Mullen, a senior software engineer at Google.
The AI agent provides access to Gemini AI software via the terminal—a text-based interface where developers input commands to control the computer. While it is geared towards aiding in code writing, the tool can be used for various tasks such as video generation or setting up a simple website.
Gemini CLI is open source, allowing users to examine the model and ensure its security. Access to the service is available from any personal Google account under a free Gemini Code Assist license.
By default, 60 requests per minute and 1,000 per day are available. With a basic paid subscription, these limits increase to 120 and 1,500, respectively.
Back in April 2024, Google introduced the Gemini Code Assist AI model for programmers.