Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: elvircrn <elvircrn@gmail.com>
Signed-off-by: Elvir Crnčević <elvircrn@gmail.com>
Co-authored-by: mgoin <mgoin64@gmail.com>
Co-authored-by: Varun Sundar Rabindranath <varunsundar08@gmail.com>
This commit is contained in:
Elvir Crnčević
2025-10-10 17:19:53 +02:00
committed by GitHub
parent ae9d0e7da5
commit 7b03584de8
6 changed files with 519 additions and 405 deletions

View File

@@ -33,11 +33,11 @@ TORCH_LIBRARY_EXPAND(TORCH_EXTENSION_NAME, ops) {
#endif
ops.def(
"silu_mul_fp8_quant_deep_gemm_cuda(Tensor input, Tensor counts, Tensor! "
"y_q, Tensor! y_s, int group_size, "
"bool use_ue8m0, int num_parallel_tokens) -> ()");
ops.impl("silu_mul_fp8_quant_deep_gemm_cuda", torch::kCUDA,
&silu_mul_fp8_quant_deep_gemm_cuda);
"persistent_masked_m_silu_mul_quant(Tensor input, Tensor counts, Tensor! "
"y_q, Tensor! y_s,"
"bool use_ue8m0) -> ()");
ops.impl("persistent_masked_m_silu_mul_quant", torch::kCUDA,
&persistent_masked_m_silu_mul_quant);
ops.def("weak_ref_tensor(Tensor input) -> Tensor");
ops.impl("weak_ref_tensor", torch::kCUDA, &weak_ref_tensor);