feat: add structured output design doc

[jimoosciuc]

CLOSES: #315

Motivation

add structured output design doc

Modifications

add structured output design doc

Accuracy Tests

only design doc

Benchmarking and Profiling

only design doc

Checklist

• Please use English, otherwise it will be closed.
• The purpose of the PR, or link existing issues this PR will resolve.
• The test plan, such as providing test command.
• (Optional) The necessary documentation update.

[gemini-code-assist]

Summary of Changes

Hello @jimoosciuc, 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 adds a foundational design document for a new structured output system. The primary goal is to enable Large Language Models to generate outputs that strictly adhere to predefined formats, such as JSON Schema, EBNF, or regular expressions. The document outlines a layered, pluggable architecture, detailing how different grammar backends will be integrated, how constraints will be applied during token generation, and the API surface for developers. It also covers performance optimizations like caching and asynchronous processing to ensure efficiency and scalability.

Highlights

• Structured Output Design Document: This pull request introduces a comprehensive design document detailing a new structured output system for LLMs, aiming to provide reliable, machine-parseable outputs.
• Pluggable Architecture: The system is designed with a four-layer pluggable backend architecture, allowing for easy integration of future grammar engines like XGrammar and Outlines, with llguidance as the default.
• Constraint Types Supported: The design supports various constraint types including JSON Schema, EBNF (Extended Backus-Naur Form), Regex, and advanced Structural Tags for tool/function calling.
• OpenAI-Compatible API: The structured output functionality is integrated into OpenAI-compatible API endpoints (/v1/chat/completions, /v1/completions) with clear syntax for specifying response formats.
• Performance Optimizations: Key performance considerations include grammar caching (compiling once, reusing many times) and an asynchronous dispatch pattern using ThreadPoolExecutor to prevent blocking, especially for potentially expensive future backend compilations.
• Mutual Exclusivity Enforcement: The system strictly enforces that only one structured constraint type can be active per request, with validation occurring at multiple layers (API, SamplingParams, and tool calling).

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. ↩