feat(core): add Gemini 3.6 Flash and 3.5 Flash-Lite model configurations

[Blackmanx]

Summary

This PR adds support and model resolution configuration for Gemini 3.6 Flash and Gemini 3.5 Flash-Lite in packages/core.

Details

• Gemini 3.6 Flash and 3.5 Flash-Lite support: Configures base model definitions, capabilities (thinking, multimodalToolUse), aliases, and Code Assist-compatible IDs for gemini-3.6-flash and gemini-3.5-flash-lite.
• Independent model resolution: Adds useGemini3_6Flash as a separate resolution condition and propagates it through direct resolution, classifier resolution, policy chains, and Auto model descriptions, preserving the existing Gemini 3.5 Flash rollout and Gemini 2.5 Flash fallback behavior.
• Internal tool optimization: Moves web helpers, loop detection, edit fixing, speaker checks, context snapshots, and Flash chat compression to Gemini 3.5 Flash-Lite. Context distillation and agent-history summarization use Gemini 3.6 Flash.
• Policy-chain integrity: Uses the explicit fallback chain Gemini 3.5 Flash-Lite -> Gemini 3.6 Flash -> Gemini 2.5 Flash -> Gemini 2.5 Pro.
• Request compatibility: Removes sampling parameters rejected or ignored by the new models after aliases and overrides are merged.
• Generated artifacts: Regenerates model-config goldens, the settings schema, and configuration reference documentation.

Related Issues

Related to #28483 and #28625.

How to Validate

npm run format
npm run build
npm run lint:ci
npm run typecheck
npm run test:ci

Expected result: every command exits successfully. The full local validation passed, including 6,957 CLI tests, 7,771 core tests, 145 A2A tests, 134 script tests, and 17 SEA-launch tests.

Pre-Merge Checklist

• Updated relevant documentation and README (if needed)
• Added or updated tests
• No breaking changes
• Validated on required platforms or methods:
  • macOS
    • npm run
    • npx
    • Docker
    • Podman
    • Seatbelt
  • Windows
    • npm run
    • npx
    • Docker
  • Linux
    • npm run
    • npx
    • Docker

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[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 introduces support for the latest Gemini 3.6 Flash and 3.5 Flash-Lite models. It updates the core model configuration, resolution logic, and internal tool settings to leverage these new models while maintaining backward compatibility and policy chain integrity. Additionally, it includes necessary adjustments to documentation, schemas, and test suites to reflect these changes.

Highlights

• New Model Support: Added support and configuration for Gemini 3.6 Flash and Gemini 3.5 Flash-Lite models in the core package.
• Independent Resolution: Implemented separate resolution logic for Gemini 3.6 Flash to ensure it operates independently of existing model chains.
• Tool Optimization: Updated internal tool configurations to utilize Gemini 3.5 Flash-Lite, optimizing performance for tasks like web search and fetch.
• Sampling Parameter Cleanup: Added logic to automatically remove unsupported sampling parameters for Gemini 3.6 Flash and 3.5 Flash-Lite to prevent invalid API requests.

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 the 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 counterproductive. 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.

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. ↩

[github-actions]

📊 PR Size: size/L

• Lines changed: 792
• Additions: +686
• Deletions: -106
• Files changed: 16

[gemini-code-assist]

Code Review

This pull request introduces support for the gemini-3.6-flash and gemini-3.5-flash-lite models, updating configurations, policies, model resolution logic, and tests. It also adds a mechanism to strip unsupported sampling parameters for these models to prevent invalid API requests. The feedback highlights a test pollution risk due to module-level global mutable state and points out two test description mismatches in models.test.ts where the descriptions were updated to reference gemini-3.6-flash but the assertions still expect gemini-3.5-flash.

[Blackmanx]

/gemini review

[gemini-code-assist]

Code Review

This pull request adds support for the new gemini-3.6-flash and gemini-3.5-flash-lite models across configuration files, policy chains, chat compression services, and processors. It also introduces a mechanism in ModelConfigService to strip unsupported sampling parameters for these models to prevent invalid API requests. However, the review feedback highlights that the newly introduced useGemini3_6Flash flag is not fully integrated, as it is missing from the signatures and implementations of the model resolution functions (resolveModel, resolveClassifierModel) and is not passed to getAutoModelDescription within getAvailableModelOptions.

[Blackmanx]

/gemini review

[gemini-code-assist]

Code Review

This pull request introduces support for the gemini-3.6-flash and gemini-3.5-flash-lite models across the codebase. The changes include updating model configuration definitions, fallback policies, alias resolution logic, and chat compression services. Additionally, a new helper method removeUnsupportedSamplingParameters was added to ModelConfigService to strip unsupported penalties and custom sampling values for these models to prevent invalid requests. Relevant tests and documentation have been updated accordingly. There are no review comments, so no further feedback is provided.

_{Note: Security Review did not run due to the size of the PR.}