Public
  1. Public

onnxruntime

Public

Commits

Tianlei WuGitHub
99f0fe3fae0
Tianlei Wu authored and GitHub committed a2b0a69dccc
Update MultiHeadAttention benchmark to test CPU (#20972)

### Description
MultiHeadAttention benchmark script only supports cuda provider right
now.
This updates the script to support testing cpu operator and ploting gpu
latency.

### Motivation and Context
Benchmark for the coming cpu flash attention.