squall/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

Update Hopper performance plot for CUTLASS 3.1 + CTK 12.1 (#967)

Browse Source
This commit is contained in:
Vijay Thakkar 2023-06-01 11:52:40 -07:00 committed by GitHub
parent 7dbf423763
commit fde824af21
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
README.md
View File

@ -87,7 +87,7 @@ Starting from CUTLASS 3.0, CUTLASS removed support for the following:
# Performance # Performance
<p align="center"><img src=media/images/cutlass-3.0-gemm-peak-performance.png></p> <p align="center"><img src=media/images/cutlass-3.1-gemm-peak-performance.png></p>
CUTLASS primitives are very efficient. When used to construct device-wide GEMM kernels, CUTLASS primitives are very efficient. When used to construct device-wide GEMM kernels,
they exhibit peak performance comparable to cuBLAS for scalar GEMM they exhibit peak performance comparable to cuBLAS for scalar GEMM

BIN
media/images/cutlass-3.1-gemm-peak-performance.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 163 KiB