docs: add initial version of lora desgin docs

[JamesBrianD]

Fix SGL-7

Motivation

The documentation mentioned in the issue was scattered across Google Docs and comments. To make it easier to review, track changes, and collaborate, I converted these documents into Markdown and added them to the repository.

Modifications

Accuracy Tests

Benchmarking and Profiling

Checklist

• Please use English, otherwise it will be closed.
• The purpose of the PR, or link existing issues this PR will resolve.
• The test plan, such as providing test command.
• (Optional) The necessary documentation update.

[gemini-code-assist]

Summary of Changes

Hello @JamesBrianD, 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 delivers the initial version of the LoRA (Low-Rank Adaptation) design documentation for the sgl-jax project. It consolidates existing design discussions and information into a structured Markdown document, providing a clear roadmap for implementing multi-LoRA serving. The document covers the architectural choices, component-level designs, and the end-to-end request flow, with a strong emphasis on how these designs are tailored to leverage JAX's unique characteristics, such as its functional programming paradigm and static compilation, to ensure efficient and scalable LoRA inference.

Highlights

• Initial LoRA Design Documentation: This pull request introduces the foundational design documentation for implementing multi-LoRA (Low-Rank Adaptation) serving within the sgl-jax project, centralizing previously scattered information.
• Comprehensive Architecture Overview: The documentation provides a detailed overview of the system architecture, module organization, data flow, and memory management strategies for LoRA, including pre-allocated memory pools and adapter loading strategies.
• JAX-Specific Considerations: A significant portion of the design addresses the unique challenges and implications of integrating LoRA with JAX's functional programming model, static compilation, PyTree requirements, and tensor parallelism.
• Core Component Design: Detailed designs for key components such as LoRAConfig, LoRAAdapter, LoRALayers (e.g., LinearWithLoRA), LoRAKernel (BGMV/SGMV), and LoRAManager are outlined, explaining their responsibilities and interactions.
• Development Work Breakdown: The document includes a phased development plan, breaking down the implementation into foundational, core infrastructure, integration, and optimization stages, complete with specific tasks and deliverables.

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. ↩