vllm-project / vllm-project/production-stack
feature: refactoring monolithic vllm router controller logic
Open
@JiangJiaWei1103 is already working on this.
Since Aug 19, 2025.
feature request
- Dominant language
- Python
- Stars
- 2.6k
- Forks
- 503
- Avg merge
- 4d 17h
- Merged PRs (30d)
- 8
Description
Describe the feature
- Refactoring current monolithic vllm router and runtime controller logic into multiple relevant components.
Why do you need this feature?
- Many new features are keep added on helm chart that also to be included in the operator logic eventually.
- But current vllm router and runtime controller logics are monolithic and share some common resource related logics that does not facilitate to add new features.
- Refactoring vllm router and runtime controller logic will help adding new features easily and make it easier to implement unit tests.
Additional context
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.