[Community] Nominate whx-sjtu as maintainer

[wangxiyuan]

Since the first release v0.13.0rc2 and v0.14.0rc1 in 2026 are released. We consider to refresh the maintainer team. I nominate whx-sjtu as the new maintainer. Nominate reason:

✅Review Quality‌:

He has completed 80+ reviews since Oct. 2025, including#pullrequestreview-3459750305,#discussion_r2453770390, #discussion_r2472268022,#discussion_r2480554911 with high quality.

✅ Sustained Contributions

He has deep understanding of ‌vLLM‌ and ‌vLLM Ascend‌ codebases and solid contributions of more than 60+ PR merged, including contributions both to vLLM and vLLM-Ascend. Especially, performance optimization, disaggregated-prefill, scheduler, accuracy bugfix and MLA related contributions are the main reason why I nominated him:

• Performance Optimization: On behalf of the Ascend computing product line, he contributed the first PR Pull Request #13 to vllm-ascend which introduce MindIE Turbo to improve performance. After that he continues to contribued many PRs to optimize performance of vLLM-Ascend including Pull Request #18, Pull request #204, Pull Request #2002 and Pull Request #1614 etc..
• Disaggregated Prefill: He construct the baseline for p2p disaggregated-prefill of v0 in Pull Request #694, which later also becomes baseline for current v1 p2p disaggregated-prefill.
• Ascend Scheduler: He develops AscendScheduler in Pull Request #543, which provides v0-style scheduling strategy in v1 engine and welcomed by many users. What's more he continues to imporve AscendScheduler in Pull Request #943, Pull Request #822.

✅Quality Contribution‌:

• Accuracy Bugfix: He manages to fix many important and urgent accuracy-related bugs including[ Pull Request #3163, Pull Request #95 and Pull Request #1498.
• PluggableLayer: He designs PluggableLayer for full layer pluggablility: [PluggableLayer][1/N] Define PluggableLayer by whx-sjtu · Pull Request #32331 · vllm-project/vllm
• MLA Refactor: During the reconstruction of the MLA part, he made a series of efforts to make the architecture of the MLA layer successfully attracted the attention of the vllm community, and made the MLA layer support custom op in Pull Request #23332.
• MoE Refactor for glm4_moe: He manages to make glm4_moe reuse SharedFusedMoE by contributing to vllm community in Pull Request #24849. This not only solves the accuracy issue of GLM4.5, but also helps with the refactor of our MoE backend.

✅Community Involvement‌:

In 2025 Q2/Q3, he leads [RFC]: Performance optimation of decode in DeepSeek Large EP situation. · Issue #2905 · vllm-project/vllm-ascend which optimizes performance of large EP situation.

He leads refactor of deepseek models and successfully delete all LLM models in vLLM-Ascend:

• [Model][1/N] Delete deepseek v2/v3 modeling codes. by whx-sjtu · Pull Request #3189 · vllm-project/vllm-ascend
• [Model][2/N] Remove deepseek_mtp modeling. by whx-sjtu · Pull Request #3561 · vllm-project/vllm-ascend
• [Model][3/N] Refactor sfa into mla and remove deepseek_v3_2.py by whx-sjtu · Pull Request #3769 · vllm-project/vllm-ascend
• vLLM version: v0.14.1
• vLLM main: vllm-project/vllm@d682094

[github-actions]

👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:‌‌

• A PR should do only one thing, smaller PRs enable faster reviews.
• Every PR should include unit tests and end-to-end tests ‌to ensure it works and is not broken by other future PRs.
• Write the commit message by fulfilling the PR description to help reviewer and future developers understand.

If CI fails, you can run linting and testing checks locally according Contributing and Testing.

[gemini-code-assist]

Code Review

This pull request nominates whx-sjtu as a new maintainer, reflecting their significant contributions to the project. The changes update the .github/CODEOWNERS file to grant ownership over the attention, ops, and sample modules, and add whx-sjtu to the list of committers in docs/source/community/contributors.md. The modifications are consistent with the provided rationale and appear to be correctly implemented. After a thorough review, I have not identified any issues of high or critical severity.

[MengqingCao]

welcome!

[LCAIZJ]

LGTM!