Skip to content

Conversation

@HDCharles
Copy link
Collaborator

@HDCharles HDCharles commented Nov 25, 2025

Summary
This test would ordinarily take too long so we only quantize the first 10 layers

SUMMARY:
adding a new e2e test

blocked by vllm-project/vllm#28878 since we need to be able to run it in vllm as part of the test

TEST PLAN:
N/A

Summary
This test would ordinarily take too long so we only quantize the first
10 layers

Signed-off-by: HDCharles <charlesdavidhernandez@gmail.com>
@HDCharles HDCharles marked this pull request as draft November 25, 2025 21:40
@github-actions
Copy link

👋 Hi! Thank you for contributing to llm-compressor. Please add the ready label when the PR is ready for review.

Note: This is required to complete the testing suite, please only add the label once the PR is code complete and local testing has been performed.

@gemini-code-assist
Copy link
Contributor

Summary of Changes

Hello @HDCharles, 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 new end-to-end regression test designed to validate the W4A16 grouped quantization scheme for the Qwen3-30B-A3B model. The test setup includes a specific configuration that limits the quantization process to the initial ten layers of the model, a strategic choice made to optimize test execution time without compromising the core validation objective.

Highlights

  • New End-to-End Test: Introduced a new end-to-end test for the Qwen3-30B-A3B model, specifically targeting W4A16 grouped quantization.
  • Quantization Scope Optimization: The test is configured to quantize only the first 10 layers of the model to reduce execution time, as noted in the PR description.
  • Configuration Files Added: Two new YAML files were added: one for the test configuration (qwen3_w4a16_grouped_quant.yaml) and another for the quantization recipe (recipe_w4a16_group_quant_first_10_layers.yaml).
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.

@HDCharles HDCharles added the ready When a PR is ready for review label Nov 25, 2025
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 adds a new end-to-end test for Qwen3 MoE model with W4A16 quantization. The changes look good, but I've found a couple of issues. There's a typo in a directory name (WNA16 instead of W4A16) in the test configuration, which will cause the test to fail. Additionally, the regex used to ignore layers during quantization is not robust and will fail for models with 100 or more layers. I've provided suggestions to fix both issues.

kylesayrs
kylesayrs previously approved these changes Nov 26, 2025
Signed-off-by: HDCharles <39544797+HDCharles@users.noreply.github.com>
Signed-off-by: HDCharles <39544797+HDCharles@users.noreply.github.com>
Signed-off-by: HDCharles <39544797+HDCharles@users.noreply.github.com>
Signed-off-by: HDCharles <39544797+HDCharles@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ready When a PR is ready for review

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants