[diffusion] CI: enable warmup as default

[mickqian]

Motivation

Modifications

Accuracy Tests

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 or contact authorized users to do so.
   • /tag-run-ci-label, /rerun-failed-ci, /tag-and-rerun-ci
4. After green CI and required approvals, ask Merge Oncalls to merge.

[gemini-code-assist]

Summary of Changes

Hello @mickqian, 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 standardizes the warmup behavior for diffusion models by making it a default setting. This change simplifies the configuration of test cases and server arguments by removing the explicit warmup parameter, ensuring that performance warmup is consistently applied across relevant tests. The associated performance baselines have also been updated to reflect this new default, providing accurate metrics under the standardized warmup conditions.

Highlights

• Default Warmup Enabled: The warmup option for diffusion models is now enabled by default, removing the need for explicit configuration in test cases and server arguments.
• Simplified Configuration: The warmup parameter has been removed from DiffusionServerArgs and various DiffusionTestCase definitions, streamlining the setup for performance testing.
• Performance Baseline Updates: Performance baselines in perf_baselines.json were updated to reflect the new default warmup behavior, including the removal of a dedicated _warmup entry for zimage_image_t2i.

🧠 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.

Changelog

• python/sglang/multimodal_gen/test/scripts/gen_perf_baselines.py
  • Modified _build_server_extra_args to unconditionally add the --warmup flag.
• python/sglang/multimodal_gen/test/server/ascend/testcase_configs_npu.py
  • Removed warmup=True from DiffusionServerArgs for the video modality test case.
• python/sglang/multimodal_gen/test/server/perf_baselines.json
  • Updated stages_ms, denoise_step_ms, expected_e2e_ms, expected_avg_denoise_ms, and expected_median_denoise_ms for zimage_image_t2i.
  • Removed the zimage_image_t2i_warmup entry entirely.
• python/sglang/multimodal_gen/test/server/test_server_common.py
  • Modified diffusion_server to unconditionally add the --warmup flag to extra_args.
• python/sglang/multimodal_gen/test/server/testcase_configs.py
  • Removed the warmup field from the DiffusionServerArgs class definition.
  • Removed warmup=True from numerous DiffusionServerArgs instances within from_req_perf_record for various models and modalities.

Activity

• No human activity (comments, reviews) has been recorded on this pull request yet.

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. ↩

[mickqian]

/tag-and-rerun-ci

[gemini-code-assist]

Code Review

This pull request makes warmup a default for diffusion model CI tests. This is achieved by removing the warmup configuration from test cases and hardcoding the --warmup flag when launching the test server. The changes are consistent across test configurations, test runners, and performance baselines.

I've added two comments with suggestions to make the enforcement of warmup more robust by handling potential conflicts with arguments passed through the extras field. This will prevent unexpected behavior and make the code more maintainable.