Support new DeepGEMM format in per token group quant (part 2: srt) #7155
Conversation
There was a problem hiding this comment.
Summary of Changes
Hello @fzyzcjy, 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 is the second part of adding support for a new DeepGEMM format within the per-token group quantization mechanism for the SGLang runtime (srt). It introduces changes to handle a specific scale format (ue8m0) by modifying the scale tensor allocation and passing a new flag to the underlying kernel.
Highlights
- DeepGEMM Scale Format Support: Added a new boolean parameter
scale_ue8m0to thesglang_per_token_group_quant_fp8function to enable support for a specific scale format likely related to DeepGEMM. - Conditional Scale Tensor Allocation: Implemented new logic to allocate the scale tensor (
x_s) with a different shape, dtype (torch.int), and permutation/slicing when thescale_ue8m0flag is true, accommodating the requirements of the new scale format.
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
-
Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configureGemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩
There was a problem hiding this comment.
Code Review
This pull request introduces support for a new DeepGEMM format. My review identified a critical issue regarding the memory layout of the x_s tensor. Other points include clarifying the source of an align function, the choice of torch.int for storing packed scales, the use of a magic number 128, and addressing a TODO for consistency.
Motivation
Modifications
Checklist