update docs and reorg /test

[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 significantly updates and reorganizes the project's documentation in both Chinese and English, making it more comprehensive and user-friendly. It also introduces new testing utilities and refactors parts of the inference backend, alongside expanding support for new models and enhancing existing benchmarking tools. The changes aim to improve clarity, maintainability, and functionality across the project.

Highlights

• Comprehensive Documentation Overhaul: I've performed a major update and reorganization of both Chinese and English documentation. This includes new guides for performance benchmarking, detailed API server parameters, DeepSeek model deployment, multimodal inference, OpenAI API compatibility, and reward model configurations.
• Quickstart Simplification: The quickstart guides have been streamlined by moving advanced deployment configurations, such as PD separation, into dedicated, more detailed documentation pages. The example model in quickstart has also been updated to Qwen3-8B.
• Expanded Model Support & Configuration: Support has been added for new models including DeepSeek-V3, Qwen2-VL, and Qwen2-Reward. Configurations for various MoE models (like Grok1, Jamba, Dbrx) have also been updated within benchmark scripts to reflect broader compatibility.
• Inference Backend Refactoring: The static inference benchmark scripts have been refactored to utilize new ModelInput and ModelOutput objects, improving code structure and maintainability for future development.
• Enhanced MoE Kernel Benchmarking: MoE kernel tuning and benchmarking scripts have been improved with added support for FP8 quantization and dynamic model configuration loading, enabling more flexible and accurate performance analysis.
• New Testing Utilities: New scripts have been introduced for comparing inference results against previous commits (accuracy testing) and for validating LightLLM's compatibility with the OpenAI API, enhancing our testing capabilities.
• Installation and Docker Updates: Docker installation instructions have been revised, including updates to shm-size examples and specific Docker images for DeepSeek EP mode. Triton installation recommendations for various GPU architectures have also been updated.

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

The pull request includes updates to documentation, including README, installation guide, quick start guide, and API parameter details. It also reorganizes the /test directory. The changes aim to improve clarity, provide more accurate information, and enhance the overall user experience.