Support vLLM v0.14.0

[pan-x-c]

Description

This PR adds support for vLLM v0.13.0 and v0.14.0

Checklist

Please check the following items before code is ready to be reviewed.

• Code has passed all tests
• Docstrings have been added/updated in Google Style
• Documentation has been updated
• Code is ready for review

[gemini-code-assist]

Summary of Changes

Hello @pan-x-c, 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 focuses on updating the system's integration with vLLM to support its latest versions, 0.13.0 and 0.14.0. It involves a significant refactoring of how the vLLM API server is initialized, introducing a more modular and version-aware approach. This ensures that the system can leverage the performance and features of newer vLLM releases while maintaining stability and correct functionality across different versions.

Highlights

• vLLM Version Compatibility: Extended support for vLLM to include versions 0.13.0 and 0.14.0, ensuring the system remains compatible with the latest advancements in the vLLM library.
• API Server Initialization Refactoring: The logic for launching the vLLM API server has been refactored into a new centralized get_api_server function. This function dynamically selects and applies the correct version-specific patch for the API server, simplifying maintenance and improving adaptability to future vLLM updates.
• New vLLM v0.13.0+ API Patch: A new patch file, api_patch_v13.py, has been introduced to specifically handle the API server implementation details for vLLM versions 0.13.0 and above. This patch includes adaptations for running the server within a Ray actor environment, addressing signal handling and socket management.
• Prompt Logprobs Patch Update: The existing patch for extracting prompt logprobs has been updated to extend its compatibility range, now supporting vLLM versions up to 0.14.0.

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 vLLM versions v0.13.0 and v0.14.0. The core change involves refactoring the API server initialization to dynamically load version-specific patches based on the detected vLLM version. A new api_patch_v13.py file has been added to handle the API changes in vLLM versions 0.13.0 and above, and existing version compatibility ranges in worker_patch.py have been updated accordingly. This approach enhances maintainability by centralizing version-dependent logic and ensures compatibility with newer vLLM releases. The changes are well-implemented and align with the stated objective of the pull request.