Add MoE Integration Tests For CUTLASS Coverage

[Jonahcb]

Motivation

Add 4 more MoE integration tests to achieve complete CUTLASS path coverage.

Modifications

Adds coverage for these 4 CUTLASS kernels: cutlass_fused_experts_fp8, cutlass_w4a8_moe, cutlass_w4a8_moe_deepep_normal, cutlass_w4a8_moe_deepep_ll.

cutlass_moe_fp4 is already covered by an existing test case. Thus, all 5 CUTLASS paths are tested.

Accuracy Tests

moe_runner_cutlass_fp8:

Score: 0.600
Writing report to /tmp/mmlu_gaunernst_DeepSeek-V2-Lite-Chat-FP8.html
{'other': np.float64(0.5), 'other:std': np.float64(0.5), 'score:std': np.float64(0.48989794855663565), 'stem': np.float64(1.0), 'stem:std': np.float64(0.0), 'score': np.float64(0.6)}

moe_runner_cutlass_w4a8:

Total latency: 18.074 s
Score: 0.600
Writing report to /tmp/mmlu_tencent_DeepSeek-V3.1-Terminus-W4AFP8.html
{'other': np.float64(0.5), 'other:std': np.float64(0.5), 'score:std': np.float64(0.48989794855663565), 'stem': np.float64(1.0), 'stem:std': np.float64(0.0), 'score': np.float64(0.6)}

moe_runner_cutlass_w4a8_deepep_normal:

Score: 0.800
Writing report to /tmp/mmlu_tencent_DeepSeek-V3.1-Terminus-W4AFP8.html
{'other': np.float64(0.75), 'other:std': np.float64(0.4330127018922193), 'score:std': np.float64(0.4000000000000001), 'stem': np.float64(1.0), 'stem:std': np.float64(0.0), 'score': np.float64(0.8)}

moe_runner_cutlass_w4a8_deepep_ll:

Total latency: 15.509 s
Score: 0.600
Writing report to /tmp/mmlu_tencent_DeepSeek-V3.1-Terminus-W4AFP8.html
{'other': np.float64(0.5), 'other:std': np.float64(0.5), 'score:std': np.float64(0.48989794855663565), 'stem': np.float64(1.0), 'stem:std': np.float64(0.0), 'score': np.float64(0.6)}

Benchmarking and Profiling

Checklist

• Format your code according to the Format code with pre-commit.
• Add unit tests according to the Run and add unit tests.
• Update documentation according to Write documentations.
• Provide accuracy and speed benchmark results according to Test the accuracy and Benchmark the speed.
• Follow the SGLang code style guidance.

Review Process

1. Ping Merge Oncalls to start the PR flow. See the PR Merge Process.
2. Get approvals from CODEOWNERS and other reviewers.
3. Trigger CI tests with comments (/tag-run-ci-label, /rerun-failed-ci, /tag-and-rerun-ci) or contact authorized users to do so.
4. After green CI and required approvals, ask Merge Oncalls to merge.

[gemini-code-assist]

Summary of Changes

Hello @Jonahcb, 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 enhances the testing infrastructure for Mixture-of-Experts (MoE) operations by introducing new integration tests. The primary goal is to achieve complete path coverage for all relevant CUTLASS kernels, thereby improving the reliability and robustness of MoE implementations within the system.

Highlights

• Expanded MoE Test Coverage: Four new integration tests have been added to specifically target Mixture-of-Experts (MoE) functionality.
• Complete CUTLASS Path Validation: These new tests ensure comprehensive coverage for four critical CUTLASS kernels: "cutlass_fused_experts_fp8", "cutlass_w4a8_moe", "cutlass_w4a8_moe_deepep_normal", and "cutlass_w4a8_moe_deepep_ll".
• Full CUTLASS MoE Kernel Testing: With the inclusion of these tests, all five CUTLASS paths for MoE are now covered, including the previously existing "cutlass_moe_fp4" test case.

🧠 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 adds four new integration tests for MoE with CUTLASS kernels, improving test coverage for cutlass_fused_experts_fp8, cutlass_w4a8_moe, cutlass_w4a8_moe_deepep_normal, and cutlass_w4a8_moe_deepep_ll. The changes are straightforward and align with the goal of complete CUTLASS path coverage. My main feedback is regarding code duplication in the test configurations, which could be refactored for better maintainability.