[Feature]: How to run speculative models with tensor parallelism?

[cxxuser]

🚀 The feature, motivation and pitch

I noticed that the current speculative mode does not support tp from this link (https://docs.vllm.ai/en/stable/models/spec_decode.html).

However, not supporting TP will greatly limit the choice of speculative models. I would like to know why there is no TP support for speculative models. I am trying to read and modify this part of the code, but I don't understand why the scorer model can support TP, but the speculative model cannot. What are the considerations in system design?

Alternatives

No response

Additional context

No response

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.

[ShangmingCai]

This issue is related to the following issue and pull.
#5814
#5414

[github-actions]

This issue has been automatically marked as stale because it has not had any activity within 90 days. It will be automatically closed if no further activity occurs within 30 days. Leave a comment if you feel this issue should remain open. Thank you!

[github-actions]

This issue has been automatically closed due to inactivity. Please feel free to reopen if you feel it is still relevant. Thank you!