vit + qwen3 moe

[shihaobai]

No description provided.

[gemini-code-assist]

Summary of Changes

Hello @shihaobai, 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 integrates the Qwen3 MOE model into the lightllm framework under the InternVL multimodal model family. My aim is to expand the range of supported large language models, particularly those with Mixture-of-Experts architectures, within the existing multimodal capabilities.

Highlights

• New Model Integration: I've added support for the Qwen3 MOE model within the InternVL multimodal framework, specifically introducing InternVLQwen3MOETpPartModel.
• Configuration Handling: The new InternVLQwen3MOETpPartModel includes a custom _init_config method to correctly load and normalize llm_config parameters from config.json, ensuring compatibility with existing lightllm conventions.
• Module Imports: Necessary imports for the new Qwen3 and Qwen3 MOE InternVL models have been added to lightllm/models/__init__.py, making them discoverable within the system.

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 introduces support for Qwen3 MOE models within the InternVL framework. It adds necessary imports and registers the new model, ensuring it can be used for multimodal tasks. The changes seem straightforward and well-structured.

[gemini-code-assist]

It seems like InternVLQwen3MOETpPartModel should be placed after InternVLQwen3TpPartModel to maintain consistency with the other models.

Suggested change

	InternVLQwen3TpPartModel,
	InternVLQwen3MOETpPartModel,
	InternVLQwen3MOETpPartModel,
	InternVLQwen3TpPartModel,

[gemini-code-assist]

Consider renaming Qwen3MOEModel to Qwen3MOETpPartModel for consistency with other models in this file.

Suggested change

	from lightllm.models.qwen3_moe.model import Qwen3MOEModel
	from lightllm.models.deepseek2.model import Deepseek2TpPartModel
	from lightllm.models.qwen3_moe.model import Qwen3MOETpPartModel

[gemini-code-assist]

The condition argument in @ModelRegistry should be checked for None to avoid potential errors if llm_model_type_is returns None.

Suggested change

	@ModelRegistry(["internvl_chat"], is_multimodal=True, condition=llm_model_type_is("qwen3_moe"))
	class InternVLQwen3MOETpPartModel(Qwen3MOEModel):
	@ModelRegistry(["internvl_chat"], is_multimodal=True, condition=llm_model_type_is("qwen3_moe") if llm_model_type_is else None)

[gemini-code-assist]

Consider renaming InternVLQwen3MOETpPartModel to InternVLQwen3MOEModel for consistency with other models in this file.

Suggested change

	class InternVLQwen3MOETpPartModel(Qwen3MOEModel):
	class InternVLQwen3MOEModel(Qwen3MOEModel):