In this talk, we'll dive into the exciting world of Gemma, a groundbreaking family of open AI models by Google. We'll explore how you can leverage Gemma's capabilities to build innovative Mobile and Web projects. One of these capabilities is running on your mobile device or browser without an internet connection.

Key takeaway:

- Understand the core functionalities and architectures of Gemma, lightweight, state-of-the-art models
- Discover the advantages of using Gemma, including its open-source nature, versatility across tasks, and on-device/browser capabilities
- Explore different approaches to fine-tuning of Gemma model, that let you achieve better results on a wide number of tasks. Once a model has been fine-tuned, you won't need to provide as many examples in the prompt.
- Discover practical methods for integrating Gemma into your Mobile and Web applications, unlocking new possibilities for user interaction and functionality
- Explore the different types of Gemma (CodeGemma, PaliGemma, etc.), the purposes for which they can be used, how Gemma differs from Gemini, and explore the distinct characteristics of each model, including size, performance, and optimal use cases
- Discover Gemma alternatives which you can use on-device and in browser

PS: Additionally I'll introduce the Flutter plugin that I created and currently maintain for the integration of Gemma into Flutter apps