Google Introduces Gemma 4 for Local Agentic AI on Android
Google has announced the integration of Gemma 4, its latest open-source AI model, into the Android ecosystem to enhance local agentic intelligence. This update targets the entire software lifecycle, from development to production, by offering two primary capabilities. First, Android Studio now supports local-first agentic coding, allowing developers to use Gemma 4 for complex reasoning, tool calling, and code refactoring directly on their machines without cloud dependency. Second, the company introduced the AICore Developer Preview, enabling developers to prototype on-device intelligence using Gemma 4 E2B and E4B models via the ML Kit GenAI Prompt API. Gemma 4 serves as the foundation for the upcoming Gemini Nano 4, which promises significant performance improvements, including speeds up to four times faster and battery usage reduced by 60 percent compared to previous versions. This initiative aims to empower developers to build sophisticated, privacy-focused AI experiences that run entirely on Android hardware, preparing apps for the release of new flagship devices later this year.
Wire timeline
Google Introduces Gemma 4 for Local Agentic AI on Android
Google has announced the integration of Gemma 4, its latest open-source AI model, into the Android ecosystem to enhance local agentic intelligence. This update targets the entire software lifecycle, from development to production, by offering two primary capabilities. First, Android Studio now supports local-first agentic coding, allowing developers to use Gemma 4 for complex reasoning, tool calling, and code refactoring directly on their machines without cloud dependency. Second, the company introduced the AICore Developer Preview, enabling developers to prototype on-device intelligence using Gemma 4 E2B and E4B models via the ML Kit GenAI Prompt API. Gemma 4 serves as the foundation for the upcoming Gemini Nano 4, which promises significant performance improvements, including speeds up to four times faster and battery usage reduced by 60 percent compared to previous versions. This initiative aims to empower developers to build sophisticated, privacy-focused AI experiences that run entirely on Android hardware, preparing apps for the release of new flagship devices later this year.
Android Developers Blog