关于深度学习推理时使用tensorrt(trt)和Triton Inference Server(Triton)的可行性探讨 #9440
1746104160
started this conversation in
Ideas
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
说说你遇到的问题?
局限性
tensorrt
Triton Inference Server
优点
tensorrt
trt是n卡专用的c++深度学习推理框架,maa项目也是基于c++的。满足在极其特殊的情况下需要快速推理的需求。
Triton Inference Server
Triton可以在n卡gpu和cpu上实现高性能高并发推理,并且有专门的Triton client库用于c++运行推理。满足在极其特殊的情况下需要快速推理的需求。
Beta Was this translation helpful? Give feedback.
All reactions