Google is rolling out a free version of its AI coding tool called Gemini Code Assistant to all users. “Developers worldwide now get free AI-assisted coding help with the highest usage limits available, as well as code review assistance,” the search giant said in a blog post published on Tuesday, February 25. Google’s AI coding tool is powered by its Gemini 2.0. The foundational AI model has been fine-tuned to carry out coding-related tasks by analysing and validating a large number of real-world coding use cases. As a result, Gemini Code Assistant can be used to generate code in any programming language. Google said that Gemini Code Assistant supports over 1,28,000 input tokens and uses natural language to generate, explain, or improve code. “For example, a freelance website developer could quickly get code with a prompt like, 'Build me a simple HTML form with fields for name, email, and message, and then add a ‘submit’ button.' Or someone just looking to automate more routine tasks can ask Gemini to 'Write a script that sends a daily email with the latest weather forecast,' or “Explain what this Python code snippet does and find any errors'," the blog post read. While there are several free coding assistants on the market, Google said that users will get “practically unlimited capacity” with up to 180,000 code completions per month using Gemini Code Assist. The free version of the AI coding tool has also been released for public preview on Microsoft-owned developer platform GitHub. Gemini Code Assist for GitHub can also be used to review existing code in both public and private repositories. Code reviews are said to be a critical part of the software development process. Google said its AI-generated code reviews are provided by AI agents within Gemini. “ Offloading basic reviews to an AI agent can help make code repositories more maintainable and improve quality, allowing developers to focus on more complex tasks,” the company said. Over 25 per cent of all new code at Google is AI-generated, as per the tech giant. However, the code is accepted only after being reviewed by human engineers. Studies have found that there has been a decline in the quality of code generated using AI tools. Last year, a survey of 800 software developers using GitHub Copilot found that none of them saw any productivity gains. The survey respondents also reported experiencing a 41 per cent increase in bugs within pull requests while using the AI coding assistant to write code.
