[Bug]: OmniModelConfig mismatch with vllm main [not yet in released vllm 0.16.0]

[divyanshsinghvi]

Your current environment

The output of python collect_env.py

Your output of `python collect_env.py` here

Your code version

The commit id or version of vllm

The commit id or version of vllm-omni

🐛 Describe the bug

vllm_omni/engine/arg_utils.py

Migrating vllm-omni to support vllm 0.16.0 missed removing

logits_processor_pattern=self.logits_processor_pattern,

but correctly added the parameter logits_processors=self.logits_processors (the new name)

Add new parameter :

language_model_only=language_model_only,

Before submitting a new issue...

• Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.

[tzhouam]

I also observed these changes in vllm yesterday, but the vllm v0.16.0 tag released 2 weeks ago may not contain these commits.

[divyanshsinghvi]

I also observed these changes in vllm yesterday, but the vllm v0.16.0 tag released 2 weeks ago may not contain these commits.

I think you are right. I was building on the top of latest. Probably we might have to wait if they release with these changes otherwise will close the issue.

[Gaohan123]

@divyanshsinghvi vllm 0.16.0 just released. Could you check if it still exists?

[divyanshsinghvi]

@Gaohan123 I will check and update.

[divyanshsinghvi]

@Gaohan123 the release cut off data is still Feb 8. So it's likely not yet there, probably on next release. There are few more issues that I saw, but will update it once the new release is there in vllm.

[tzhouam]

This change maybe included in v0.17.0 release and already included in the rebase branch.
Closed and see if retains in the v0.17.0rc1