Skip to content

shauray8/cutlass_q8gemm

Repository files navigation

cutlass Q8 GEMM

PyTorch fp8 precision gemm lib w/ cutlass, this is inspired by aredden/torch-cublas-hgemm transfering its HGEMM to cutlass and get FP8 GEMM working with cutlass

figure out how to get nf4 linear layers to cutlass

About

PyTorch fp8 precision gemm lib w/ cutlass

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published