Google Unveils Gemini Nano’s Power for Third-Party Android Apps

Expanding AI Capabilities with On-Device Execution Google is set to unleash the full potential of its Gemini Nano AI model for third-party Android apps, following a significant expansion of its availability and introduction of image input support. The move marks a major step forward in the development of on-device generative AI technology, which is poised…

Editorial Staff

May 17, 2025

1–2 minutes

AI, AICore, Android, APIs, Apps, Gemini, Gemini Nano, Google

Expanding AI Capabilities with On-Device Execution

A hand holding a smartphone displaying a digital representation of AI technology, symbolizing the expansion of on-device generative AI capabilities.

Google is set to unleash the full potential of its Gemini Nano AI model for third-party Android apps, following a significant expansion of its availability and introduction of image input support.

The move marks a major step forward in the development of on-device generative AI technology, which is poised to revolutionize the way we interact with digital content.

According to a recent session list for I/O 2025, Google will introduce a new set of generative AI APIs that harness the power of Gemini Nano. These APIs will enable developers to implement use cases such as summarizing, proofreading, and rewriting text, as well as generating image descriptions.

The new APIs will be built on top of the AICore, allowing for on-device execution of AI foundation models like Gemini Nano.

In a move that suggests Google is preparing to formally announce the new APIs, the company has quietly published documentation on ML Kit’s new GenAI APIs. The GenAI APIs will provide developers with a simplified, high-level interface to harness the power of Gemini Nano, enabling them to deliver out-of-the-box performance for common tasks such as summarization, proofreading, and image description generation.

Compared to the existing AI Edge SDK, ML Kit’s GenAI APIs will be offered in “beta” instead of “experimental access,” indicating that Google may allow apps to use the new APIs in production.

The GenAI APIs will also support images, enabling the image description feature, and will be available on any Android phone that supports the multimodal Gemini Nano model, including devices from manufacturers such as HONOR, Motorola, OnePlus, Samsung, and Xiaomi.

The expansion of Gemini Nano’s capabilities is expected to have significant implications for the development of AI-powered apps, enabling developers to create more sophisticated and user-friendly experiences.

With the new GenAI APIs, developers will be able to tap into the power of Gemini Nano to deliver high-performance AI features, while keeping data processing local and enhancing user privacy.