Alphabet Inc.’s Google is adding an artificial intelligence coding assistant to ease the work of developers, aiming to catch up with rival products such as OpenAI’s Codex and Anthropic’s Claude Code.
The Gemini CLI—short for “command line interface”—is launching globally, the internet search giant said on 25 June. The tool is designed to reduce the complexity of traditional programming interfaces, letting developers use ordinary language to work with AI.
“With Gemini CLI, you can have a natural language conversation with your computer to solve problems, to weave complex workflows together to do way more than you could have possibly done in the past,” Taylor Mullen, a Google senior staff software engineer, said in a briefing with reporters ahead of the company’s announcement.
The AI agent provides access to Google’s Gemini AI software from the terminal, the text-based interface where developers type commands to control their computer. Although the tool is primarily an AI coding assistant, developers can use Gemini CLI for a range of tasks, including generating video or setting up a simple website, Google said.
Gemini CLI is also open source, meaning its code is freely available to modify and redistribute. The company said that its goal is to democratize AI coding and continually improve the product with users’ contributions. Developers can also inspect the code to understand how it works and verify its security, Google said.
The company’s TensorFlow AI engine is also open source, as are several AI models based on its so-called transformers. Those are the building blocks that make up today’s most widely used large language models, including OpenAI’s ChatGPT.
In 2024, Google introduced a pair of open large language models that it called Gemma, reversing its general strategy of keeping the company’s proprietary artificial intelligence technology out of public view. The announcement on Wednesday points to a renewed commitment to more actively engage with outside developers.
Users can access Gemini CLI from any personal Google account, which will grant them a free Gemini Code Assist license. The license grants access to Gemini 2.5 Pro, the company’s flagship model, as well as what Google says is the industry’s largest allowance of prompts—the queries that trigger the AI.
Users can make 60 requests per minute and 1,000 requests per day. Paying for a standard version increases that allowance to 120 queries per minute and 1,500 queries per day, while Gemini CLI’s enterprise model allows for 2,000 queries per day.
“We believe that these tools are going to dominate the way not just developers, but creators of all kinds, work over the next decade,” said Ryan J. Salva, a senior director of product at Google.
“It doesn’t matter whether you’re a student or a hobbyist, a freelancer or a developer. You should have access to the same tools.”
2025 Bloomberg L.P. Distributed by Tribune Content Agency, LLC.
Citation:
Google debuts Gemini AI coding tool in bid to entice developers (2025, June 26)
retrieved 26 June 2025
from https://techxplore.com/news/2025-06-google-debuts-gemini-ai-coding.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.
