add --enable_multimodal_visual --enable_multimodal_audio start args.

[hiworldwzj]

No description provided.

[gemini-code-assist]

Summary of Changes

Hello, 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 enhances the flexibility of enabling multimodal capabilities within the LightLLM server by introducing distinct command-line arguments for visual and audio modules. This change allows users to activate visual (--enable_multimodal_visual) and audio (--enable_multimodal_audio) functionalities independently, moving away from a single, overarching --enable_multimodal flag. The underlying system now dynamically recognizes and initializes the appropriate multimodal components, improving modularity and configuration precision for diverse multimodal AI applications.

Highlights

• Granular Multimodal Control: Introduced dedicated command-line arguments --enable_multimodal_visual and --enable_multimodal_audio for granular control over multimodal module activation.
• Refactored Startup Logic: Refactored the application startup logic to automatically set the general enable_multimodal flag if either visual or audio multimodal features are explicitly enabled.
• Independent Service Routing: Updated the HTTP server manager to independently establish connections and route requests to visual and audio multimodal services based on their respective enablement flags.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Changelog

• lightllm/server/api_cli.py
  • Renamed --enable_multimodal to --enable_multimodal_visual and updated its help message.
  • Clarified the help message for --enable_multimodal_audio.
• lightllm/server/api_start.py
  • Added logic to set args.enable_multimodal = True if args.enable_multimodal_visual or args.enable_multimodal_audio is true.
  • Decoupled the starting of visual and audio processes, making them conditional on their respective enable_multimodal_visual and enable_multimodal_audio flags.
• lightllm/server/core/objs/start_args_type.py
  • Added enable_multimodal_visual as a boolean field to the StartArgs class.
• lightllm/server/httpserver/manager.py
  • Modified the HttpServerManager constructor to conditionally create send_to_visual and send_to_audio ZMQ sockets based on args.enable_multimodal_visual and args.enable_multimodal_audio respectively.
  • Updated transfer_to_next_module to route requests to send_to_visual or send_to_audio based on the specific multimodal flags.

Activity

• No specific activity was provided for this pull request.

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 refactors the multimodal support by splitting the single --enable_multimodal flag into two more specific flags: --enable_multimodal_visual and --enable_multimodal_audio. This allows for more granular control over which multimodal features are enabled. The changes are consistent across the argument parsing, server startup logic, and request handling. The implementation appears correct. I have one minor suggestion to improve code readability.

[gemini-code-assist]

For better readability and to make the control flow clearer, it's better to use an elif here. Since each if block ends with a return, these conditions are mutually exclusive in execution. Using elif makes this explicit.

Suggested change

	if self.args.enable_multimodal_visual:
	self.send_to_visual.send_pyobj(group_req_objs.to_group_req_index(), protocol=pickle.HIGHEST_PROTOCOL)
	return
	if self.args.enable_multimodal_audio:
	self.send_to_audio.send_pyobj(group_req_objs.to_group_req_index(), protocol=pickle.HIGHEST_PROTOCOL)
	return
	if self.args.enable_multimodal_visual:
	self.send_to_visual.send_pyobj(group_req_objs.to_group_req_index(), protocol=pickle.HIGHEST_PROTOCOL)
	return
	elif self.args.enable_multimodal_audio:
	self.send_to_audio.send_pyobj(group_req_objs.to_group_req_index(), protocol=pickle.HIGHEST_PROTOCOL)
	return