Google unveils Gemini Code Assist, available to all developers at no cost.

Google has launched an exciting new AI coding helper called Gemini Code Assist, which anyone can download and use for free. This tool is based on a special version of Google’s advanced language model and comes packed with many useful features.
Gemini Code Assist arrives as the race in the AI field gets more intense, especially among tech companies and research labs that are working hard on AI advancements. Recently, another coding assistant named Claude 3.7 Sonnet was also introduced, showing that many companies are investing in this technology.
Gemini Code Assist is powered by a version of Google’s Gemini model called Gemini 2.0. Although we don’t have all the details on how this model works or the specific data used for its training, we know that it has been fine-tuned with real coding examples. The model is privately held, so we can only make educated guesses about its inner workings.
What sets Gemini Code Assist apart is its ability to handle a large volume of information, with a context window of 128,000 tokens. This means it can efficiently manage tasks that involve multiple files, making it especially useful for developers. The full version of the Gemini models can handle up to 2 million tokens, but coding tasks are often better suited to smaller context sizes since they require pulling together relevant information from various parts.
Accessing Gemini Code Assist is easy. Google has incorporated this AI helper into many of its online applications, including popular development tools like Colab, Android Studio, and Firebase. It is also available in various integrated development environments (IDEs) such as Visual Studio Code and JetBrains’ PyCharm. To start using it, all you need to do is install an extension for your chosen IDE and log in with your Google Account. The free version allows you 6,000 requests each day and a total of 180,000 requests each month, which should be more than sufficient for most developers.
In your IDE, there are many ways to benefit from Gemini Code Assist. You can generate new code, ask for explanations of existing code, and even create unit tests for specific functions. Additionally, as you write code, the tool provides helpful autocomplete suggestions based on the context of your project and your notes, enhancing the overall coding experience.
Google has also made Gemini Code Assist available for GitHub, which assists users in reviewing code, making changes, managing pull requests, and overseeing repositories.
Alongside the free version, there are also Standard and Enterprise editions for more demanding users and organizations that need greater capabilities.
Personally, I am quite impressed with the developments Google has made in the AI space over the recent months. They have made significant strides to catch up with other leading AI companies like OpenAI and Anthropic. With their extensive resources—both in terms of data and finance—I believe Google is well-positioned to lead in the area of generative AI tools very soon.