Google LLC today unveiled more features for Android Studio, its toolkit for building Android apps, powered by its Gemini artificial intelligence model.

Gemini in Android Studio rolled out as an AI upgrade for Android Studio, a type of software editor, in May. It brought Google’s flagship model to the development tool, providing developers the ability to chat with the AI model, receive insights about Android software, and generate code suggestions.

“Gemini goes beyond just guidance. It can edit your code, helping you quickly move from prototype to implementation, implement common design patterns, and refactor your code,” said Sandhya Mohan, product manager of Android Studio. “Gemini also streamlines your workflow with features like documentation and commit message generation, allowing you to focus more time on writing code.”

According to Google, with the new updates, users will be able to go beyond generating code and modify code directly in Android Studio using prompts according to custom needs. For example, a developer could select a code block and type, “Improve the readability of my code” or “Change this code to accept an extra variable for indicating customer name.”

Naming classes, methods and variables is probably the worst part of every developer’s day, aside from writing documentation, so now Gemini can assist with generating intuitive names. This function can trigger while developers code or work as part of refactoring code to streamline readability. The AI can also generate code documentation with a simple right-click.

When it comes time to commit changes to version control, the AI can examine the changes done to the code and suggest a simple summary of the developer’s work. This makes it quicker and easier to do commits and reduces the number that have badly documented or blank commit messages. That way, if the code needs to be reverted to a previous state, a coworker can easily understand why the changes were made before rolling them back.

Developers will also be able to save and mange their most commonly used prompts using a prompt library. With a single click, they can recall them from context menu and use them again.

In a big update, Gemini can use its multimodal image capabilities to help developers visualize user interface elements during design time in Android Studio. Manually building mock data into UI elements can be difficult and time-consuming, but Gemini can now auto-generate previews for Composable elements with relevant context using AI to simplify that process.

Google said the company is continuing to experiment with Gemini’s multimodal support to help with UI development. In a future update, Gemini in Android Studio will allow developers to attach custom images for context to make it easier for the AI to create beautiful and engaging UI interfaces.

According to Google, AI integration doesn’t stop with coding, it also extends into release. By integrating AI into the App Quality Insights tool, developers can analyze crashes reported by Google Play Console and Firebase Crashlytics.

New features for code quality also include unit test scenario generation based on local code context. This will allow developers to quickly test and discover flaws in code before it’s shipped. If the code fails to build or synchronize during the compile phase, Google said Gemini will also assist with understanding errors and provide insights on how to fix the code.

These new features are available to try today in the Android Studio early access canary channel. Most of the features will be rolling out in the upcoming Ladybug feature drop, which is expected to be released in late December.

Google ups speed of Android software kit releases in 2025

Next year, Google will have more frequent Android software development kit releases. The company said the faster rollouts would help provide developers and users with higher stability and drive innovation for apps on its devices.

This means there will be a major release in the second quarter and a minor release in the fourth quarter, both of which will include new developer application programming interfaces. The first release will include behavioral changes that can affect apps, so developers may want to adjust and test code afterward. In between, incremental quarterly releases will include feature updates and optimizations.

Image: Google