[NVIDIA] Support nvfp4 cutlass gemm (#13571)
This commit is contained in:
@@ -152,6 +152,11 @@ torch::Tensor ggml_mul_mat_a8(torch::Tensor W, torch::Tensor X, int64_t type,
|
||||
int64_t row);
|
||||
|
||||
#ifndef USE_ROCM
|
||||
void cutlass_scaled_fp4_mm(torch::Tensor& D, torch::Tensor const& A,
|
||||
torch::Tensor const& B, torch::Tensor const& A_sf,
|
||||
torch::Tensor const& B_sf,
|
||||
torch::Tensor const& alpha);
|
||||
|
||||
bool cutlass_scaled_mm_supports_fp8(int64_t cuda_device_capability);
|
||||
bool cutlass_scaled_mm_supports_block_fp8(int64_t cuda_device_capability);
|
||||
|
||||
|
||||
Reference in New Issue
Block a user