Edgar Survints / Android Authority
tl; drag
- Google is increasing access to the new ML Kit Geneai APIS, Gemini Nano, its on -device AI model.
- This new APIS, which will be announced in I/O 2025, will easily enable developers to easily implement features such as text smooth, proof reading, rewriting, and image detail generation in their apps.
- Unlike the experimental AI Edge SDK, MLC’s Geni APis beta will be in the support image input, and will be available on a wide range of Android devices beyond the Pixel 9 series.
Generative AI technology is changing how we communicate and create online content. Many people ask Google Gemini to perform tasks such as AI Chat Boats to summarize an article, e -mail -proof reading, or rewrite a message. However, some people are cautious to use these AI chat boats, especially when these tasks include highly personal or sensitive information. Google, Google, offers a small, better version of its AI model to address these concerns of privacy, Gemini Nano, which runs directly on the device instead of a cloud server. Although access to Gemini Nano is still limited to a single device line and only text only input, Google will soon increase its availability and introduce image input support.
Last month, Google published a session list for I/O 2025, which includes a session, titled, “Gemini Nano on Android: On the Device General AI with the building.” The session describes that it will “introduce a new set of Generative AI APIs that will use the power of Gemini Nano. These new APIS make it easy to implement, proof readings, and re -write the text, as well as to prepare image explanations.”
In October, Google opened experimental access to Gemini Nano through AI Edge SDK, allowing third -party developers to experience with a text to text prompt on Pixel 9 series. The AI Edge SD text -based features such as refrigerating, smart responses, proof readings, and samarizations, but does not include specific support to create image details, Google has been highlighted for the upcoming I/O session. Thus, it is likely that the “New Set of Generative AI AIPIS” in the description of the session refers to something different from the AI Edge SDK or its new version. Fortunately, we don’t have to wait until next week to find out.
Earlier this week, Google quietly published documents on the new MLC’s new Geni APIS. The ML Kit is an SD that allows developers to take advantage of machine learning capabilities in their apps without the need to understand that. The new Ginai APIS developers “allow Gemini to use the power of Nano to provide out -of -the -box performances for normal tasks through an easy, high -level interface.” Like the AI Edge SD, it is also “built on aero”, which is “implementing the on -device of the AI Foundation models like Gemini Nano, keeping data processing localized and increasing the use of the app and increasing the privacy of the user.”
Mashal Rehman / Android Authority
In other words, the Geni APIS developers of the ML Kit make it easy to use Gemini Nano privately and with high performance for different features in their apps. These features include currently summarizing, proofing, or rewriting the text, as well as adding image specifications. What is mentioned in the description of the I/O session in these four features shows that Google plans to formally announce the MLC’s Geni APIS next week.
Here is a summary of all the features offered by the Geni APIS of ML Kit:
- Abstract: Summarize articles or chat conversations as a list of tablets.
- Produces up to three tablet points
- Languages: English, Japanese and Korean
- Proof readingsShort material by improving grammar and fixing spelling mistakes.
- Languages: English, Japanese, German, French, Italian, Spanish and Korean
- Write again: Re -write short chat messages in different heads or styles.
- Styles: wide, emigrants, short, friendly, professional, referred
- Languages: English, Japanese, German, French, Italian, Spanish and Korean
- Image detail: Prepare a brief explanation of a given image.
Compared to the current AI Edge SDK, ML Kit’s geni APIS will be presented in “beta” instead of “experimental access”. The position of this ‘beta’ may mean that Google will allow apps to use new Genai APIS in production. Currently, developers cannot issue apps using AI Edge SDK, which means that no third party apps can take advantage of Gemini Nano at this time. Another difference is that AI Age SD’s text is limited to the input, while ML Kit supports Geni APIS images. This photo collaboration enables the image detail feature, which allows apps to produce a brief description of any icon.
The biggest difference between the current version of the AI Edge SDK and ML KIT’s Geneai APIS, though, is device support. Although AI Edge SD only supports Google Pixel 9 series, ML Kit’s Geni APIS can be used on any Android phone that supports the multi -modal Gemini Nano model. This includes devices such as Honor Magic 7, Motorola Razer 60 Ultra, One Plus 13, Samsung Galaxy S25, Xiaomi 15, and more.
Developers, who are interested in trying Gemini Nano in their apps, can start reading public documents for ML Kit Geneai APIS.
Have a tip? Talk to us! Email our staff at News@Androidauthority.com. You can remain anonymous or get the credit for information, it’s your choice.
