FastDeploy
FastDeploy copied to clipboard

Published 20 hours ago •

Reame
Issues

[LLM] refine llm infer

Open juncaipeng opened this issue 5 months ago • 0 comments

PR types(PR类型)

Description

decouple docker and source files
add basic tests and benchmark
refine triton and engine module

Sep 25 '24 05:09 juncaipeng