Google integrates Gemini Pro large-scale language models into Android Studio

Google LLC today rolled out upgrades to the Coding Assistant in Android Studio, its toolkit for building Android apps.

Android Studio is a so-called IDE, or integrated development environment. It is a type of word processor that allows developers to write code for new apps. Android Studio also provides other features, such as an emulator that lets you run virtual Android handsets on your Windows machine and use them to test new code.

Last May, Google added an artificial intelligence coding assistant called Studio Bot to the IDE. It was based on a large-scale language model called PaLM 2 that the search giant announced in the same month. Today’s update replaces Studio Bot in Android Studio with Gemini, an improved coding assistant based on new AI models.

Six months after the debut of PaLM 2 last year, Google announced its new flagship LLM lineup. This series, called the Gemini series, includes three of his models with different feature sets and hardware requirements. Android Studio’s upgraded chatbot is based on Gemini Pro, the mid-range model in the series, and is designed to balance response quality with hardware efficiency.

“If you were already using Studio Bot in the Canary channel, you’ll continue to experience the same useful and powerful features, but you’ll notice an improved quality of response compared to previous versions.” , Sandhya Mohan, Android Studio Product Manager, explains more in the following article: blog post today.

Gemini in Android Studio can answer common questions about Android, such as how to integrate your app with the operating system’s location tracking features. Chatbots can also answer project-specific questions. For example, a developer might ask for advice on how to remove duplicate records from a dataset that the app uses to power some functionality.

Gemini in Android Studio can also perform more complex tasks. According to Google, chatbots can rewrite code files in a new programming language and generate entire interface sections of an app. Gemini in Android Studio can also create documentation that explains how newly written code works.

“Gemini is only available if you log in and enable it,” Mohan elaborates. “You don’t need to send code context to take advantage of most features. By default, Gemini chat responses in Android Studio are based purely on conversation history and share additional context for customized responses. You can control whether you do it or not.”

The chatbot is based on Gemini 1.0 Pro, an LLM that Google introduced last December. Since then, the search giant has detailed a new version of the model called Gemini 1.5 Pro. The latter model is likely to be rolled out to Android Studio in a future update, given the fact that it has enhanced coding capabilities.

According to Google, Gemini 1.5 Pro exceeded Android Studio uses this version 1.0 in 87% of benchmarks evaluated by engineers. In one internal test, the search giant provided Gemini 1.5 Pro with a prompt containing over 100,000 lines of code. They were found to be more proficient at both generating explanations and suggesting improvements.

image: Google