diff --git a/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/fp8_blockwise_gemm.hip b/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/fp8_blockwise_gemm.hip
index ca44bba425..53b8020c6b 100644
--- a/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/fp8_blockwise_gemm.hip
+++ b/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/fp8_blockwise_gemm.hip
@@ -15,8 +15,6 @@
 #include <c10/hip/HIPStream.h>
 #include <torch/torch.h>
 
-#if defined(USE_ROCM)
-
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
 #include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
@@ -272,5 +270,3 @@ at::Tensor f8f8bf16_blockwise(
 }
 
 } // namespace fbgemm_gpu
-
-#endif // defined(USE_ROCM)
diff --git a/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/fp8_rowwise_gemm.hip b/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/fp8_rowwise_gemm.hip
index b076e281e6..ca3eba949b 100644
--- a/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/fp8_rowwise_gemm.hip
+++ b/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/fp8_rowwise_gemm.hip
@@ -18,8 +18,6 @@
 #include <c10/hip/HIPStream.h>
 #include <torch/torch.h>
 
-#if defined(USE_ROCM)
-
 #include "kernels/fp8_rowwise_kernel_manifest.h"
 
 namespace fbgemm_gpu {
@@ -302,5 +300,3 @@ at::Tensor f8f8bf16_rowwise(
 }
 
 } // namespace fbgemm_gpu
-
-#endif // defined(USE_ROCM)
diff --git a/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/fp8_tensorwise_gemm.hip b/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/fp8_tensorwise_gemm.hip
index 04b2ba3b04..108c6075e3 100644
--- a/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/fp8_tensorwise_gemm.hip
+++ b/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/fp8_tensorwise_gemm.hip
@@ -15,8 +15,6 @@
 #include <c10/hip/HIPStream.h>
 #include <torch/torch.h>
 
-#if defined(USE_ROCM)
-
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
 #include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
@@ -243,5 +241,3 @@ at::Tensor f8f8bf16_tensorwise(
 }
 
 } // namespace fbgemm_gpu
-
-#endif // defined(USE_ROCM)
diff --git a/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/kernels/fp8_rowwise_common.h b/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/kernels/fp8_rowwise_common.h
index 7177205802..caa7e5e2c0 100644
--- a/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/kernels/fp8_rowwise_common.h
+++ b/fbgemm_gpu/experimental/gen_ai/src/quantize/ck_extensions/kernels/fp8_rowwise_common.h
@@ -15,8 +15,6 @@
 #include <c10/cuda/CUDAStream.h>
 #include <torch/torch.h>
 
-#if defined(USE_ROCM)
-
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
 #include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
@@ -200,5 +198,3 @@ at::Tensor f8f8bf16_rowwise_impl(
 
   return Y;
 }
-
-#endif // defined(USE_ROCM)