Skip to content

fix vllm embedding #5413

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Aug 17, 2025
Merged

fix vllm embedding #5413

merged 2 commits into from
Aug 17, 2025

Conversation

tastelikefeet
Copy link
Collaborator

PR type

  • Bug Fix
  • New Feature
  • Document Updates
  • More Models or Datasets Support

PR information

Write the detail information belongs to this PR.

Experiment results

Paste your experiment result here(if needed).

Copy link
Contributor

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Summary of Changes

Hello @tastelikefeet, 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 addresses a bug related to VLLM embedding functionality. The core change involves enhancing the robustness of how PoolingParams are initialized within the VLLM inference engine, specifically for embedding tasks. By implementing a try-except block, the system can now adapt to different vllm library versions or API specifications, ensuring that embedding requests are processed correctly whether the task argument is supported or not. This fix aims to improve the stability and compatibility of the VLLM embedding pipeline.

Highlights

  • Robust PoolingParams Instantiation: The _add_request method, specifically when handling embedding task types, now attempts to instantiate vllm.pooling_params.PoolingParams with task='embed'.
  • Error Handling for PoolingParams: A try-except TypeError block has been added around the PoolingParams instantiation. If passing task='embed' raises a TypeError (e.g., due to an older vllm version not supporting this argument), it gracefully falls back to instantiating PoolingParams() without arguments.
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. 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.

Copy link
Contributor

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request addresses a bug in vLLM embedding by handling different versions of PoolingParams for backward compatibility. The current implementation uses a try-except block. While functional, I've suggested an alternative approach using inspect.signature to check for parameter existence, which is more explicit and consistent with the coding style in other parts of the file. This will improve code clarity and maintainability.

Comment on lines 311 to 314
try:
pooling_params = PoolingParams(task='embed')
except TypeError:
pooling_params = PoolingParams()
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

While using a try-except block for version compatibility is a valid approach (EAFP), it can sometimes obscure the programmer's intent and potentially catch unrelated TypeError exceptions. A more explicit and arguably cleaner method, which is already used elsewhere in this file for checking API capabilities, is to use inspect.signature. This directly checks if the PoolingParams constructor accepts the task argument, making the code more readable and robust against unexpected errors.

Suggested change
try:
pooling_params = PoolingParams(task='embed')
except TypeError:
pooling_params = PoolingParams()
parameters = inspect.signature(PoolingParams).parameters
if 'task' in parameters:
pooling_params = PoolingParams(task='embed')
else:
pooling_params = PoolingParams()

@tastelikefeet tastelikefeet merged commit a06d47f into modelscope:main Aug 17, 2025
1 of 2 checks passed
Jintao-Huang pushed a commit that referenced this pull request Aug 17, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants