Add support for SGLang generate API to compare and match GPT-OSS reference implementation in v6.0 (part of #4)

[nv-alicheng]

What does this PR do?

Modularizes worker.py to more easily add support for multiple APIs, adds support for SGLang's /generate API endpoint.

#4

Tested locally running GPT-OSS on H200x8.

Type of change

• Bug fix
• New feature
• Documentation update
• Refactor/cleanup

Related issues

Testing

• Tests added/updated
• All tests pass locally
• Manual testing completed

Checklist

• Code follows project style
• Pre-commit hooks pass
• Documentation updated (if needed)

[github-actions]

MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅

[gemini-code-assist]

Summary of Changes

Hello @nv-alicheng, 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 significantly enhances the system's flexibility by integrating support for the SGLang /generate API, alongside the existing OpenAI API. The core change involves modularizing the API interaction layer, allowing for dynamic selection of API types and streamlining the process of adding new API endpoints in the future. This refactoring ensures that different inference endpoints can be benchmarked and probed using their native API specifications, improving compatibility and maintainability.

Highlights

• SGLang API Support: Added comprehensive support for SGLang's /generate API endpoint, allowing the system to interact with SGLang-compatible inference servers.
• API Type Selection: Introduced a new --api-type command-line argument for probe and benchmark commands, enabling users to specify whether to use openai (default) or sglang API types.
• Modular API Adapters: Refactored the worker.py module and HTTP client configuration to dynamically select API adapters (e.g., OpenAIMsgspecAdapter, SGLangGenerateAdapter) based on the configured api_type, improving modularity and extensibility for future API integrations.
• Streaming Response Accumulators: Implemented OpenAISSEAccumulator and SGLangSSEAccumulator classes to handle streaming responses from different API types, correctly processing and accumulating Server-Sent Events (SSE) chunks.
• Dedicated Type Definitions: Extracted OpenAI-specific msgspec structs into a new openai/types.py file and created new sglang/types.py and sglang/adapter.py files to define SGLang API request/response structures and the corresponding adapter logic.

🧠 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.

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 support for the SGLang /generate API and refactors worker.py to better accommodate multiple API types. The changes are well-structured, with the introduction of an APIType enum and dedicated SSE accumulator classes being a significant improvement for modularity. However, I've identified a critical bug in the SGLangSSEAccumulator that prevents streaming from working as intended. Additionally, I have a couple of medium-severity suggestions to enhance documentation clarity and code maintainability. Once the critical bug is addressed, this will be a solid contribution.

[arekay-nv]

Amazing. Thanks!