News for hackers. Stuff that matters.

A comment on CUTLASS, a CUDA C++ template library for matrix multiply on GPUs. Parent comment

Re:CUTLASS, a CUDA C++ template library for matrix multiply ... by bsprings on Wednesday December 06, 2017 @04:33PM (#15865221)
For most formats it achieves >90% of the performance of the hand-tuned assembly implementations in cuBLAS.
[ Parent | on HN ]

There are no comments on this comment yet.