[Misc] change example code default serve port#1588
[Misc] change example code default serve port#1588lengrongfu wants to merge 1 commit intovllm-project:mainfrom
Conversation
Signed-off-by: rongfu.leng <lenronfu@gmail.com>
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 64d5f5ac2d
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
| "--port", | ||
| type=int, | ||
| default=8091, | ||
| default=8000, |
There was a problem hiding this comment.
This default switch to 8000 breaks the current qwen2.5 example workflow that still brings up the Omni server on 8091 and invokes this client without passing --port (see tests/examples/online_serving/test_qwen2_5_omni.py, where test_params uses port 8091 and commands only set --query-type). In that path the client now targets localhost:8000, causing connection failures and invalidating the example tests/documented default behavior.
Useful? React with 👍 / 👎.
| "--port", | ||
| type=int, | ||
| default=8091, | ||
| default=8000, |
There was a problem hiding this comment.
Changing this default to 8000 also breaks qwen3 flows that assume 8091: the Omni test fixture starts the server on 8091 and calls this script without a --port override (tests/examples/online_serving/test_qwen3_omni.py, test_params plus command construction). Those runs now hit localhost:8000 instead of the running server, so the example CI path fails unless every caller is updated in the same change.
Useful? React with 👍 / 👎.
1 participant
PLEASE FILL IN THE PR DESCRIPTION HERE ENSURING ALL CHECKLIST ITEMS (AT THE BOTTOM) HAVE BEEN CONSIDERED.
Purpose
Currently, the Omni service is started using
vllm serve Qwen/Qwen3-Omni-30B-A3B-Instruct --omni, which uses port 8000 by default. However, some examples use port 8091 as the default.Therefore, it would be helpful to change the default port from 8091 to 8000 in the examples.
Test Plan
Test Result
Essential Elements of an Effective PR Description Checklist
supported_models.mdandexamplesfor a new model. Please runmkdocs serveto sync the documentation editions to./docs.BEFORE SUBMITTING, PLEASE READ https://github.com/vllm-project/vllm-omni/blob/main/CONTRIBUTING.md (anything written below this line will be removed by GitHub Actions)