[Kernel] Add w8a8 CUTLASS kernels (#4749)

This commit is contained in:
Tyler Michael Smith
2024-05-16 18:32:50 -04:00
committed by GitHub
parent 8435b207af
commit 2060e93659
10 changed files with 1197 additions and 2 deletions
+8
View File
@@ -155,6 +155,14 @@ torch::Tensor gptq_marlin_repack(
int64_t size_k,
int64_t size_n,
int64_t num_bits);
int cutlass_scaled_mm_dq(
torch::Tensor& out,
torch::Tensor const &a,
torch::Tensor const &b,
torch::Tensor const &a_scales,
torch::Tensor const &b_scales);
#endif
void squeezellm_gemm(