In this talk, we’ll explore the groundbreaking Gemma family of open AI models by Google and how to leverage their capabilities for Mobile and Web projects. Discover how these lightweight, state-of-the-art models can run directly on devices or browsers without an internet connection.

Key takeaways:

Understand Gemma’s core functionalities, architectures, and open-source advantages.
Learn how Gemma excels in versatility across tasks and enables on-device/browser capabilities.
Explore fine-tuning techniques for Gemma to achieve better task results with fewer prompt examples.
Gain insights into integrating Gemma into Mobile and Web apps, unlocking new interaction possibilities.
Compare Gemma variants (CodeGemma, PaliGemma, etc.), their use cases, and how they differ from Gemini in size, performance, and applications.
Discover alternatives like Llama for on-device or in-browser use.
PS: I’ll also introduce the Flutter plugin I developed to integrate Gemma into Flutter apps seamlessly.