Google has quietly released Google AI Edge Gallery, an app that allows users to run a range of openly available AI models from the AI dev platform Hugging Face on their phones. While the app is currently available only on Android, it is yet to release on iOS, according to media reports.
Edge Gallery runs offline, and doesn’t require an internet connection. It allows users to find, download, and run compatible models that generate images, answer questions, write and edit code, and more.
Google calls this “an experimental Alpha release.” The app can be downloaded from GitHub and displays shortcuts to AI functions such as image analysis or chatting on the home screen. Selecting a function brings up a list of suitable models, including Google’s Gemma 3n. It also offers a “prompt lab” which allows users to start a task with a single input.
READ: Meta releases Llama 4 AI models to rival DeepSeek and OpenAI (
The prompt lab contains templates and adjustable parameters to customize the behavior of the models. Users can choose a language model themselves via Hugging Face integration, or use their own language model via LiteRT.
While AI models running in the cloud are often more powerful than those that run locally, they also have certain downsides. Some users may have concerns while sending sensitive data to remote data centers. They may also want to make use of AI models without an internet connection.
Google has cautioned users that their experiences may vary when it comes to performance. Newer devices with more powerful hardware will invariably run models faster, but the model size also factors into performance with larger devices taking more time to carry out tasks than smaller ones.
Google is also currently inviting developers to give feedback on the Google AI Edge Gallery experience. The app is under an Apache 2.0 license, meaning it can be used in most contexts without restriction.