feat: Add JSON Schema support for controlled generation

[gericdong]

Add JSON Schema support for controlled generation

Also re-organize the code to make the flow more clearer.

[gemini-code-assist]

Summary of Changes

Hello @gericdong, 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 enhances the controlled generation capabilities of the Gemini API by adding direct support for JSON Schema. The accompanying notebook has been thoroughly reorganized to provide a clearer, more comprehensive guide on generating structured outputs, including JSON and Enum values, using various schema definitions.

Highlights

• New Feature: JSON Schema Support: This pull request introduces explicit support for JSON Schema in controlled generation, allowing users to define and enforce structured outputs from the Gemini API using the JSON Schema standard. This is demonstrated via the new response_json_schema parameter.
• Notebook Reorganization and Clarity: The intro_controlled_generation.ipynb notebook has been significantly restructured to improve clarity and flow. It now clearly separates and details three distinct options for generating JSON (Pydantic models, OpenAPI Schema, and JSON Schema) before moving on to Enum generation and general use cases.
• Updated Examples and Model Version: Existing examples for Pydantic and OpenAPI schemas have been updated to use a more consistent 'CountryInfo' theme. A new dedicated example for JSON Schema has been added. Additionally, the notebook now defaults to using the gemini-2.5-flash model for demonstrations.

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 in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.

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 issue 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 or fill out our survey to provide feedback.

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 JSON Schema support for controlled generation and reorganizes the notebook for clarity. The new examples effectively demonstrate the features. A suggestion was made to simplify a schema in a use case example to enhance intuitiveness.

[gemini-code-assist]

The response_schema for analyzing product reviews defines an array of arrays of objects. This structure is a bit unusual. A more natural structure would be a single array of objects, where each object represents the analysis of one review. Consider simplifying it to a single list of review analysis objects.

response_schema = {
    "type": "ARRAY",
    "items": {
        "type": "OBJECT",
        "properties": {
            "rating": {"type": "INTEGER"},
            "flavor": {"type": "STRING"},
            "sentiment": {
                "type": "STRING",
                "enum": ["POSITIVE", "NEGATIVE", "NEUTRAL"]
            },
            "explanation": {"type": "STRING"}
        },
        "required": ["rating", "flavor", "sentiment", "explanation"]
    }
}