Signed-off-by: Xin Yang <xyangx@amazon.com>
This commit is contained in:
@@ -132,12 +132,6 @@ TORCH_LIBRARY_EXPAND(TORCH_EXTENSION_NAME, m) {
|
||||
// DeepSeek V3 optimized router GEMM for SM90+
|
||||
m.def("dsv3_router_gemm(Tensor! output, Tensor mat_a, Tensor mat_b) -> ()");
|
||||
// conditionally compiled so impl registration is in source file
|
||||
|
||||
// gpt-oss optimized router GEMM kernel for SM90+
|
||||
m.def(
|
||||
"gpt_oss_router_gemm(Tensor! output, Tensor input, Tensor weights, "
|
||||
"Tensor bias) -> ()");
|
||||
m.impl("gpt_oss_router_gemm", torch::kCUDA, &gpt_oss_router_gemm);
|
||||
#endif
|
||||
}
|
||||
|
||||
|
||||
Reference in New Issue
Block a user