[Misc] Qwen MoE model supports LoRA

[jeejeelee]

Essential Elements of an Effective PR Description Checklist

• The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
• The test plan, such as providing test command.
• The test results, such as pasting the results comparison before and after, or e2e results
• (Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model.

Purpose

Currently we do not support FusedMoE LoRA inference, so we add relevant warnings for users.

Fix #18120 (comment)

Test Plan

Test Result

(Optional) Documentation Update

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

🚀

[gemini-code-assist]

Summary of Changes

Hello @jeejeelee, 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 extends LoRA support to Qwen Mixture-of-Experts (MoE) models by explicitly marking them as LoRA-compatible. It also integrates a mechanism to detect the presence of FusedMoE layers within a model and provides a clear warning to users about the current lack of support for LoRA inference on FusedMoE expert weights, guiding them on how to proceed with LoRA fine-tuning.

Highlights

• LoRA Compatibility for Qwen MoE Models: I've updated the Qwen2MoeForCausalLM and Qwen3MoeForCausalLM classes to inherit from SupportsLoRA, formally declaring their compatibility with LoRA within vLLM's framework.
• Fused MoE LoRA Inference Warning: I've introduced a new utility function is_moe_model to detect if a model contains FusedMoE layers. If detected, a warning is now issued to users, clarifying that vLLM currently does not support LoRA inference for FusedMoE expert weights. This helps guide users on current limitations and best practices.

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 is currently in preview and 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 to provide feedback.

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 adds a warning for LoRA on MoE models with FusedMoE layers, as this is currently unsupported. It also enables LoRA for Qwen2 and Qwen3 MoE models. The changes look good and align with the PR's goal. I've added a couple of suggestions to improve code clarity and maintainability regarding the new warning mechanism.

[gemini-code-assist]

The attribute self.is_moe_model is initialized here, seemingly just to trigger the warning inside is_moe_model. If this attribute is not intended for future use within the LoRAModelManager class, it would be clearer to call is_moe_model without assigning its result to an attribute, indicating that only its side effect (the warning) is desired. For example:

is_moe_model(self.model)

If you do plan to use self.is_moe_model later, please consider adding a comment explaining its purpose.

[hmellor]

LGTM