[misc] Add Ray Serve to requirements to support multi-node training (#318)

This PR adds Ray Serve to the requirements to enable support for
multi-node training. It addresses the issue described here:
https://github.com/volcengine/verl/issues/87#issuecomment-2659493418

Co-authored-by: Yu Feng <fengyufengyu@didiglobal.com>
This commit is contained in:
Yu Feng
2025-02-21 13:26:43 +08:00
committed by GitHub
parent 0a1b16f800
commit 0268917a95

View File

@ -12,7 +12,7 @@ peft
pyarrow>=15.0.0
pybind11
pylatexenc
ray
ray[data,train,tune,serve]
tensordict<0.6
transformers
vllm<=0.6.3