DeepSpeed-MII
DeepSpeed-MII copied to clipboard
What is the recommended way of bringing up mii as a service
My understanding is that we have to build a fastAPI wrapper, and during intialized phase we call client = mii.client("mistralai/Mistral-7B-v0.1")
and we implement a handler to call client.generate
.
You can use the RESTFul API
Also, I added a PR #317 (still in progress) to implement an OpenAI-compatible RESTful API