Fix bias handling in TritonMoeQuantInfo within quantization/mxfp4.py#10579
Fix bias handling in TritonMoeQuantInfo within quantization/mxfp4.py#10579zhyncs merged 2 commits intosgl-project:mainfrom
Conversation
yhyang201
requested review from
BBuf,
Edwardf0t1,
HaiShaw,
Ying1123,
ch-wan,
ispobock,
kushanam,
merrymercy and
zhyncs
as code owners
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Summary of Changes
Hello @yhyang201, 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 addresses a critical field mismatch in the TritonMoeQuantInfo constructor within the MoE quantization module. By refactoring how optional bias parameters are accessed and passed, the change ensures that the system robustly handles layers both with and without bias attributes, preventing runtime errors and maintaining the integrity of the quantization process without altering existing behavior for cases where biases are absent.
Highlights
- Bias Parameter Handling: Corrected a field mismatch when constructing
TritonMoeQuantInfoby changing howw13_weight_biasandw2_weight_biasare passed. - Error Prevention: The fix now uses
getattr(layer, "attribute_name", None)to safely retrieve bias parameters, preventing potentialTypeErrorexceptions if the bias attributes are not present on the layer object. - Parameter Mapping: The
w13_weight_biasandw2_weight_biasparameters are now correctly mapped to theb13andb2fields ofTritonMoeQuantInforespectively, ensuring proper wiring of optional biases in the MoE quantization path.
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
-
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. ↩
There was a problem hiding this comment.
Code Review
This pull request addresses a bug in quantization/mxfp4.py by correcting the keyword arguments used to instantiate TritonMoeQuantInfo. The change replaces the incorrect w13_weight_bias and w2_weight_bias with the expected b13 and b2 parameters. Furthermore, the use of getattr with a default None value is a robust way to handle optional bias attributes, preventing potential errors and ensuring that biases are correctly passed when present. The fix is correct, well-contained, and improves the reliability of the MoE quantization path.
* origin/qwen3: (30 commits) chore: bump sgl-kernel 0.3.11 (sgl-project#10630) feat: add fused moe config for Qwen3-Next-80B-A3B-Instruct on B200 (sgl-project#10631) model support: Sarashina2VisionForCausalLM (sgl-project#10632) [Performance] Qwen3-Next: speed up update_mamba_state_after_mtp_verify by 10x; e2e up to 3.54% faster (sgl-project#10586) [Performance] Qwen3-Next: replace arange to cached query_start_loc_li… (sgl-project#10553) [Feature] Speculative decoding support lookahead (sgl-project#9873) refactor: use registry for _get_attention_backend_from_str (sgl-project#10629) [router] refactor worker to builder pattern 1/n (sgl-project#10628) Garbage collector regression in the online server (sgl-project#10621) feat: Add FlexAttention Backend for Efficient Sparse Attention (sgl-project#9947) Fix bias handling in TritonMoeQuantInfo within quantization/mxfp4.py (sgl-project#10579) [Performance] qwen3-next improve causal conv1d in prefill phase (sgl-project#10595) Fix sgl_kernel import failure on devices other than CUDA (sgl-project#10610) support qwen3-next-fp8 deepep (sgl-project#10622) update deepep version for qwen3-next deepep moe (sgl-project#10624) Feat/add heartbeat mechanism for nixl conn (sgl-project#10222) [RL] Add destroy process group api (sgl-project#9979) fix deepep assert when PD disaggregation == null (sgl-project#8274) Scale kkt after reduction (sgl-project#10604) [improvement] add average input/output token length for hicache benchmark stats output (sgl-project#10525) ...
4 participants
Motivation
This PR fixes a field mismatch when constructing
TritonMoeQuantInfoinquantization/mxfp4.py.w13_weight_biasandw2_weight_bias, butTritonMoeQuantInfodefines these as optional fieldsb13andb2. This could raise a TypeError and/or leave biases unwired.b13andb2, sourcing them from the layer withgetattr(..., None)to safely handle layers without bias parameters.quantization/mxfp4.py.Modifications
Accuracy Tests
Benchmarking and Profiling
Checklist