Google Launches Gemini CLI: A Powerful Agentic Tool to Challenge OpenAI and Anthropic in Developer Workflows

Google unveiled Gemini Cli, a new agentic AI tool designed to directly bring the power of its Gemini models to developers’ terminals.

Unlike Codex CLI D’OPENAI, which operates in a more limited environment, and the Code Claude d’Anthropic, which emphasizes security and a context window of 200,000 tocorides, Gemini Cli stands out with its Apache 2.0 open source license and no initial paid mass, promoting community innovation.

The open-source utility allows users to execute natural language commands for tasks such as code writing, debugging, creating features and rationalization of workflows from the command line. However, although its free level is robust, business users requiring features such as parallel agents or specific governance may need a paid API key.

Now available in preview, GEMINI CLI locally connects to the bases of developers and offers transparent access to Gemini 2.5 Pro. It has a context window of 1 million tokens.

Speaking on the launch, the CEO of Google Cloud Thomas Kurian said,

“Today, we have introduced Gemini Cli, an Open Source agent that allows developers a versatile local utility that can be directly accessible from the terminal. From code to content, in -depth research for task management, Gemini CLIs are built for your multiplate compatibility.

Gemini Cli is part of Google’s wider effort to win the Mindshare developer and promote his AI models in programming environments. He joins tools such as Gemini Code Assist and the asynchronous coding assistant Jules, but Gemini Cli marks a more direct competitor to tools based on terminals such as Codex CLA D’OPENAI and the Claude of Anthropic.

Beyond the coding, Gemini Cli extends his capacities to other areas of productivity. Developers can use it for:

• Create videos via Google’s VEO 3 model
• Generate detailed reports using the research agent in depth
• Respond to real -time data via Google search
• Connect to external databases via MCP servers

With the launch of Gemini Cli, several users praised its accessibility and power.

A user on x @rishabincloud wrote,

“Gemini Cli is here!” An Open Source agent that brings the power of Gemini directly to your terminal. With unmatched use limits for free excited to try it! Especially due to 60 model requests per minute and 1,000 requests per day at no cost. ”

@MervinPraire wrote,

“I just tried the Gemini Cli of Google and I am stunned. Building a complete URL shortcut with an prompt. Window window 1M context means that you can feed your whole code base, and it understands all that could be a serious competitor of the Claude code”

@Ben Jacobson wrote,

“Ok, so the Ga Gemini is very good. Could even be better than the release of the Claude code. And faster. Really wild moments in which we live ”.

As part of its deployment strategy and to encourage adoption, Google is Gémini CLI Open-Source under the Apache 2.0 license, which is generally considered to be one of the most permissive. The company says it expects a network of developers to contribute to the Github project. In addition, the technology giant offers generous use limits: up to 60 model requests per minute and 1,000 per day at no cost.

Since the launch of Gemini 2.5 Pro in April, Google AI models have become a favorite among the developers. With the growing popularity of Gemini 2.5 PRO among developers, already causing the use of tools like Github Copilot and Cursor, Google positions Gemini Cl Cl as a central tool to deepen direct engagement with the developer community.

The command line interface provides a rationalized and high performance gateway to take advantage of Gemini’s abilities through various technical workflows. The Google movement signals an aggressive thrust to dominate the workflows of developers, by taking advantage of free access and integration to challenge competitors.