feat(modelAvailabilityService): integrate model availability service into backend logic

[adamfweidman]

Summary

This PR refactors the model fallback mechanism to be driven by the Model Availability Service. This introduces a more robust, policy-based approach for handling model failures. All functionality is gated by an experimental flag.

Details

The previous fallback logic was tightly coupled with specific model names and error types. By integrating the Model Availability Service, we can now define flexible policies for model selection, retries, and fallbacks in centralized way.

• geminiChat.ts:
  • Integrates selectModelForAvailability to choose the best available model based on policy.
  • Dynamically updates generateContentConfig if the selected model differs or changes during retries.
• retry.ts:
  • retryWithBackoff now accepts a getAvailabilityContext callback to apply state transitions (e.g., terminal, transient) based on failure kind.
  • Marks models as healthy upon success.
• handler.ts:
• Splits logic into legacy and policy-driven paths (gated by isModelAvailabilityServiceEnabled).
• Fallback intents like retry_always now use config.setActiveModel().

Related Issues

Fixes: https://github.com/google-gemini/maintainers-gemini-cli/issues/1079

How to Validate

npm run preflight

Pre-Merge Checklist

• [ ] Updated relevant documentation and README (if needed)
• [ ] Added/updated tests (if needed)
• [ ] Noted breaking changes (if any)
• [ ] Validated on required platforms/methods:
  • [ ] MacOS
    • [x] npm run
    • [ ] npx
    • [ ] Docker
    • [ ] Podman
    • [ ] Seatbelt
  • [ ] Windows
    • [ ] npm run
    • [ ] npx
    • [ ] Docker
  • [ ] Linux
    • [ ] npm run
    • [ ] npx
    • [ ] Docker

[gemini-code-assist[bot]]

Summary of Changes

Hello @adamfweidman, I'm Gemini Code Assist[^1]! 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 introduces a significant refactoring of the model fallback mechanism, transitioning from a tightly coupled, error-type-specific approach to a more robust, policy-driven system powered by a new Model Availability Service. This change centralizes model selection, retries, and fallbacks, allowing for flexible policies to manage model health and ensure a smoother user experience, all gated by an experimental flag.

Highlights

• Policy-Driven Model Fallback: The core model fallback mechanism has been refactored to be driven by a new Model Availability Service, enabling flexible, policy-based model selection and retry strategies.
• Dynamic Model Selection & Configuration: The system can now dynamically select the best available model based on defined policies and update the generation configuration mid-retry if a fallback occurs, ensuring a smoother user experience.
• Enhanced Error Classification: A new errorClassification.ts module provides a standardized way to categorize API errors (e.g., terminal, transient, not_found), which informs the model availability policies.
• Centralized Model State Management: The Config class now includes an _activeModel property and methods (setActiveModel, resetTurn) to manage the currently active model and interact with the availability service.
• Integration into Core Components: The Model Availability Service is integrated into BaseLlmClient, GeminiClient, and GeminiChat to ensure consistent policy application across different content generation flows.

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 :thumbsup: and :thumbsdown: 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.

[^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.

[github-actions[bot]]

Size Change: +8.83 kB (+0.04%)

Total Size: 21.5 MB

Filename	Size	Change
./bundle/gemini.js	21.5 MB	+8.83 kB (+0.04%)

ℹ️ View Unchanged

Filename	Size
./bundle/sandbox-macos-permissive-closed.sb	1.03 kB
./bundle/sandbox-macos-permissive-open.sb	890 B
./bundle/sandbox-macos-permissive-proxied.sb	1.31 kB
./bundle/sandbox-macos-restrictive-closed.sb	3.29 kB
./bundle/sandbox-macos-restrictive-open.sb	3.36 kB
./bundle/sandbox-macos-restrictive-proxied.sb	3.56 kB

_{compressed-size-action}

[adamfweidman]

/gemini review