[Fix] Revert back to using CUTLASS mm_fp4 backend

[b8zhong]

Motivation

After manually installing nvidia-cudnn-cu13==9.18.0.77, it still seems to be worse (mm_fp4 picks this when cudnn >= 9.15.xx and CUDA 13, so hold off on this change for now (otherwise API will pick incorrect backend)

Issue reported in flashinfer-ai/flashinfer#2375

Modifications

Change it back to earlier behaviour.

[gemini-code-assist]

Summary of Changes

Hello @b8zhong, 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 aims to resolve a performance degradation issue encountered when using FlashInfer's FP4 GEMM operations, specifically with nvidia-cudnn-cu13. By modifying the backend selection logic, it ensures that the CUTLASS mm_fp4 backend is explicitly utilized, thereby restoring the expected performance and preventing the API from selecting a suboptimal backend.

Highlights

• FP4 GEMM Backend Reversion: This PR reverts the FP4 GEMM backend selection logic to explicitly use the CUTLASS backend for FlashInfer operations when the backend is set to auto-detection.
• Performance Regression Fix: The change addresses a reported performance regression observed with nvidia-cudnn-cu13==9.18.0.77, where the mm_fp4 backend was performing worse than expected.

🧠 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.

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 reverts the mm_fp4 backend to use 'cutlass' by default when the backend is set to 'auto'. This is a sensible temporary fix to address the performance degradation caused by flashinfer's auto-selection logic picking a suboptimal backend. The change is clear and directly addresses the issue described. I have one suggestion to improve maintainability.

[b8zhong]

/tag-and-rerun-ci

[Fridge003]

/rerun-stage unit-test-backend-4-gpu-b200

[github-actions]

✅ Triggered unit-test-backend-4-gpu-b200 to run independently (skipping dependencies).

[github-actions]

🔗 View workflow run

[Fridge003]

/rerun-stage unit-test-backend-4-gpu-b200

[github-actions]

✅ Triggered unit-test-backend-4-gpu-b200 to run independently (skipping dependencies).

[github-actions]

🔗 View workflow run

[Fridge003]

B200 fp4 gemm tests passed