[Kernel]: Cutlass 2:4 Sparsity + FP8/Int8 Quant Support (#10995)

Co-authored-by: Faraz Shahsavan <faraz.shahsavan@gmail.com>
Co-authored-by: ilmarkov <markovilya197@gmail.com>
Co-authored-by: Rahul Tuli <rahul@neuralmagic.com>
Co-authored-by: rshaw@neuralmagic.com <rshaw@neuralmagic.com>
This commit is contained in:
Dipika Sikka
2024-12-18 09:57:16 -05:00
committed by GitHub
co-authored by Faraz Shahsavan ilmarkov Rahul Tuli rshaw@neuralmagic.com <rshaw@neuralmagic.com>
parent f04e407e6b
commit 60508ffda9
30 changed files with 2365 additions and 117 deletions
+9
View File
@@ -162,6 +162,15 @@ void cutlass_scaled_mm_azp(torch::Tensor& out, torch::Tensor const& a,
torch::Tensor const& azp_adj,
c10::optional<torch::Tensor> const& azp,
c10::optional<torch::Tensor> const& bias);
void cutlass_scaled_sparse_mm(torch::Tensor& out, torch::Tensor const& a,
torch::Tensor const& b, torch::Tensor const& e,
torch::Tensor const& a_scales,
torch::Tensor const& b_scales,
c10::optional<torch::Tensor> const& bias);
bool cutlass_sparse_compress_entry(torch::Tensor& a_compressed,
torch::Tensor& e, torch::Tensor const& a);
#endif
void static_scaled_int8_quant(torch::Tensor& out, torch::Tensor const& input,