π₯
testing edge cases
TLDR; π¨π¦π§π¨πΌβπ»ποΈπ§ββοΈ
Lots of #FP16 & #BLAS these days
Interests: #AVX512 #SYCL #F1 #HPC & making code FAST
Born below 365 PPM
Stars
1
star
written in Cuda
Clear filter
FP64 equivalent GEMM via Int8 Tensor Cores using the Ozaki scheme