Android is enhancing its development tools with the introduction of Gemma 4, a new open-source AI model. Gemma 4 is designed with improved reasoning and tool-calling abilities to enhance the entire Android software lifecycle. Developers can utilize Gemma 4 locally within Android Studio for code assistance, improving development. This includes refactoring code, creating whole apps, and iterative fixes, with the model staying on the user's machine. Gemma 4 also powers the next generation of Gemini Nano, optimized for use on Android devices. The on-device model provides faster performance and uses less battery than its predecessor while operating directly on device hardware. Developers can prototype these capabilities through the AICore Developer Preview using the ML Kit GenAI Prompt API. This local-first approach promotes privacy and cost-effectiveness for developers. It offers a way to build AI functionalities into apps without relying on external cloud resources. Further development will add Gemma 4 to Android Bench for performance analysis.