Google has released the AI Edge Gallery application, which allows you to use artificial intelligence directly on your smartphone, without the Internet and without transferring data to the cloud. With this application, you can perform tasks such as image analysis or text creation using only your device's hardware resources.
What does this mean?
Privacy: All calculations take place on the device, and your data does not leave the phone. This is an ideal solution for those who work with confidential information.
Offline: No need to be connected to the internet. AI works even in remote locations or while traveling.
Savings: The app does not require subscriptions and does not consume quotas. Everything runs on your phone's computing power, which is especially useful for small businesses and hobbyists.
How does the app work?
AI Edge Gallery works with Google Gemma 3n models that are downloaded to your phone. These models autonomously perform tasks using the knowledge they gained before graduation. Models can take up from 554 MB to 4.4 GB, depending on the size.
When the model is loaded, it does not require additional data and can execute queries directly on the device. At the same time, the results are not always lightning-fast, but they are quite acceptable for basic tasks.
What are the disadvantages?
Limited model support: The application only supports the format.task, which limits the choice of available models.
Battery load: Models, especially larger ones, require a lot of computing resources, which can quickly drain the battery.
Lower power compared to cloud solutions: For more complex tasks, it is better to use cloud services.
Is it worth using?
If privacy and the ability to work offline are important to you, this app can be a useful tool. However, for more complex tasks and faster performance, cloud services such as ChatGPT are still the preferred choice.
Which is more important to you: autonomy and privacy or the power of cloud AI?