Add hybrid SDK sample

[lethargicpanda]

New AI sample leveraging the Firebase Hybrid SDK:

[gemini-code-assist]

Summary of Changes

Hello, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request integrates a new 'Gemini Hybrid' sample into the AI Sample Catalog. The sample is designed to illustrate how to leverage both on-device and cloud-based Gemini models for generative AI tasks, specifically for generating and translating hotel reviews. It provides a practical example of implementing a robust AI solution that can adapt to different device capabilities and network conditions by intelligently switching between local and remote inference.

Highlights

• New Hybrid AI Sample Added: A new sample application, 'Gemini Hybrid', has been introduced to demonstrate a hybrid approach to generative AI inference. This sample showcases the use of both on-device (Gemini Nano via ML Kit) and cloud-based (Gemini Flash via Firebase AI SDK) models, with a fallback mechanism to the cloud when on-device capabilities are unavailable or for more complex tasks.
• Hybrid Inference Capabilities: The sample allows users to generate hotel reviews based on selected topics and then translate them into various languages. It provides options to choose different inference modes (e.g., only on-device, only in cloud, prefer on-device, prefer in cloud) to observe the behavior of hybrid model deployment.
• Dependency Updates: New Firebase AI on-device dependencies (firebase-ai-ondevice and firebase-ai-ondevice-interop) have been added to libs.versions.toml to support the hybrid inference functionality.
• UI and ViewModel Implementation: The new sample includes a complete Compose UI (GeminiHybridScreen.kt) for user interaction and a ViewModel (GeminiHybridViewModel.kt) to manage the application logic, state, and interaction with the Firebase AI SDK for both generation and translation tasks.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Changelog

• app/build.gradle.kts
  • Added a project dependency for the new 'gemini-hybrid' sample.
• app/src/main/java/com/android/ai/catalog/domain/SampleCatalog.kt
  • Imported the new GeminiHybridScreen.
  • Opted into PublicPreviewAPI for the sampleCatalog.
  • Added a new SampleCatalogItem entry for the Gemini Hybrid sample, including its title, description, route, entry screen, tags, and Firebase requirement.
• app/src/main/res/values/strings.xml
  • Added new string resources for the Gemini Hybrid sample's title and description.
• gradle/libs.versions.toml
  • Added a new version reference for firebaseAiOndevice.
  • Added new library definitions for firebase-ai-ondevice and firebase-ai-ondevice-interop.
  • Updated the firebase-ai library definition to use a version reference.
• samples/gemini-hybrid/.gitignore
  • Added a .gitignore file to exclude build artifacts.
• samples/gemini-hybrid/README.md
  • Added a README file detailing the Gemini Hybrid sample's description, how it works, and key code snippets for on-device and cloud inference.
• samples/gemini-hybrid/build.gradle.kts
  • Added a new Gradle build file for the 'gemini-hybrid' module, configuring Android library settings, dependencies (including Firebase AI, ML Kit, and Compose), and Kotlin options.
• samples/gemini-hybrid/consumer-rules.pro
  • Added an empty consumer-rules.pro file.
• samples/gemini-hybrid/proguard-rules.pro
  • Added a proguard-rules.pro file with standard ProGuard comments.
• samples/gemini-hybrid/src/main/java/com/android/ai/samples/geminihybrid/GeminiHybridScreen.kt
  • Added the Compose UI screen for the Gemini Hybrid sample, including components for topic selection, inference mode dropdown, review generation, translation, and status display.
• samples/gemini-hybrid/src/main/java/com/android/ai/samples/geminihybrid/GeminiHybridViewModel.kt
  • Added the ViewModel for the Gemini Hybrid sample, managing UI state, inference mode, selected tags, review text, and implementing the logic for generating and translating content using Firebase AI SDK.
• samples/gemini-hybrid/src/main/res/values/strings.xml
  • Added specific string resources for the Gemini Hybrid sample, including titles, descriptions, button texts, language options, status messages, and review topics.
• settings.gradle.kts
  • Included the new ':samples:gemini-hybrid' project in the build.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request introduces a new 'Gemini Hybrid' sample, demonstrating a hybrid approach with on-device and cloud-based models. The implementation is well-structured within the new module, utilizing a ViewModel and Jetpack Compose for the UI.

My review focuses on adherence to the repository's architecture guidelines, code consistency, and maintainability. Key feedback includes suggestions to align with the recommended layered architecture, improve composable function signatures for better reusability, and correct some inconsistencies in the build configuration and documentation. Addressing these points will enhance the quality and clarity of the new sample.

[lethargicpanda]

cc @calren @JolandaVerhoef