Want to develop Android apps using Gemini Nano? We've got some good news
Gemini Nano excels in on-device tasks
2 min. read
Published on
Share this article
Improve this guide
Read our disclosure page to find out how can you help MSPoweruser sustain the editorial team Read more
Key notes
- Gemini Nano is now available for Android developers to build apps with on-device AI using the AI Edge SDK.
- It allows private, offline processing for tasks like rephrasing, proofreading, and summarization.
- Developers can experiment on Pixel 9 devices with customizable inference parameters.
Since Google launched Gemini Nano, its AI model that’s more efficient for on-device tasks, developers have been eager to use the model for Android apps.
The good news now is that Google has opened up access to the model for all Android app developers via AI Edge SDK, which was previously available only through a closed preview. Initially, you can test text-to-text prompts only on Pixel 9 devices, with plans to expand support to additional devices and modalities soon.
Gemini Nano is the lightweight sibling of the Gemini family. As a model that excels in on-device tasks, Nano eliminates the reliance on cloud servers. That means, app developers can use this model instead for apps with tasks like text generation and rephrasing.
“Since on-device generative AI models run on devices with less computational power than cloud servers, they are significantly smaller and less generalized than their cloud-based equivalents,” Google boasts.
Google Pixel 9 lineup was launched back in August this year, consisting of Pixel 9, Pixel 9 Pro, and Pixel 9 Pro XL powered by the new Tensor G4 chip. It started hitting the shelves in early September in the US starting at $799.
Earlier in April, we reported on core findings in the AICore app and Google Messages beta, which hinted at several AI smarts for Google Pixel 9. These features include Optical Character Recognition (OCR), text-to-image generation, smart replies, and more.
User forum
0 messages