[Misc]: Ask for the roadmap of async output processing support for speculative decoding

[Lin-Qingyang-Alec]

Anything you want to discuss about vllm.

I wonder when speculative decoding can support async output processing, it is really a important feature.

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.