Last week, Google quietly released an application that allows users to perform a series of open AI models available from the AI Dev platform that hugs the face on their phones.
Called Google Ai Edge GalleryThe app is available for Android and will soon come to iOS. It allows users to find, download and perform compatible models that create images, answer questions, write and process the code and much more. The models run offline, without the need to connect the internet, utilizing the processors of the supported phones.
AI models running in the cloud are often more powerful than their local counterparts, but also have their disadvantages. Some users may be cautious about sending personal or sensitive data to a remote data center or want to have models without having to find a Wi-Fi or Cellular connection.
Google Ai Edge Gallery, which Google calls a “Alpha Experimental Release”, can be downloaded from gitHub following These instructions. The home screen shows shortcuts in tasks and capabilities such as “Ask Image” and “Ai Chat.” Running in a skill draws a list of work -like models, such as Google’s Gemma 3N.
The Google AI Edge Gallery also provides a “instant lab” that users can use to start “Single-Turn” work powered by models, such as summarizing and re-registration. The immediate workshop comes with various task standards and adjustable settings to refine models’ behaviors.
Your kilometers may differ in terms of performance, Google warns. Modern devices with more powerful material will perform predictable models faster, but the size of the model is also important. Larger models will take more time to complete a task – let’s say, answering a question about a picture – from the smaller models.
The attractive members of the Google Community of developers to provide feedback on the Google AI Edge Gallery experience. The application is under Apache 2.0 license, which means that it can be used in most frames – commercially or otherwise – without restriction.
