diff --git a/library/src/tensor_operation_instance/gpu/gemm_multi_abd/CMakeLists.txt b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/CMakeLists.txt
index 7c22d86810..5af7322b1a 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multi_abd/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/CMakeLists.txt
@@ -2,9 +2,14 @@
 set(GEMM_MULTI_ABD_INSTANCES)
 
 list(APPEND GEMM_MULTI_ABD_INSTANCES 
+	device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
+	device_gemm_xdl_multi_abd_bias_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
+	device_gemm_xdl_multi_abd_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
 	device_gemm_xdl_multi_abd_bias_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
 	device_gemm_xdl_multi_abd_bias_gelu_bf16_i8_bf16_mk_nk_mn_v1_instance.cpp
-
+	device_gemm_xdl_multi_abd_multiply_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
+	device_gemm_xdl_multi_abd_multiply_bias_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
+	device_gemm_xdl_multi_abd_multiply_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
 	device_gemm_xdl_multi_abd_multiply_bias_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
 	)
 
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
new file mode 100644
index 0000000000..573dcc7d76
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
@@ -0,0 +1,58 @@
+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved.
+
+#include <cstdlib>
+
+#include "ck/ck.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_abd_xdl_cshuffle.hpp"
+
+#include "device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_common.hpp"
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+void add_device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_v1_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleABD<AsLayout,
+                                                      ck::Tuple<B0Layout, B1Layout>,
+                                                      ck::Tuple<>,
+                                                      ELayout,
+                                                      AsDataType,
+                                                      ck::Tuple<B0DataType, B1DataType>,
+                                                      ck::Tuple<>,
+                                                      EDataType,
+                                                      AElementOp,
+                                                      Multiply,
+                                                      PassThrough>>>& instances)
+{
+    add_device_operation_instances(instances,
+                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_comp_instances<
+                                       ck::Tuple<B0Layout, B1Layout>,
+                                       ck::Tuple<>,
+                                       ck::Tuple<B0DataType, B1DataType>,
+                                       ck::Tuple<>,
+                                       Multiply,
+                                       PassThrough,
+                                       GemmMNKPadding,
+                                       Interwave>{});
+    add_device_operation_instances(instances,
+                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_mem_instances<
+                                       ck::Tuple<B0Layout, B1Layout>,
+                                       ck::Tuple<>,
+                                       ck::Tuple<B0DataType, B1DataType>,
+                                       ck::Tuple<>,
+                                       Multiply,
+                                       PassThrough,
+                                       GemmMNKPadding,
+                                       Interwave>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_bias_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_bias_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
new file mode 100644
index 0000000000..6833ab20f8
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_bias_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
@@ -0,0 +1,58 @@
+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved.
+
+#include <cstdlib>
+
+#include "ck/ck.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_abd_xdl_cshuffle.hpp"
+
+#include "device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_common.hpp"
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+void add_device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_bias_v1_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleABD<AsLayout,
+                                                      ck::Tuple<B0Layout, B1Layout>,
+                                                      ck::Tuple<D0Layout>,
+                                                      ELayout,
+                                                      AsDataType,
+                                                      ck::Tuple<B0DataType, B1DataType>,
+                                                      ck::Tuple<D0DataType>,
+                                                      EDataType,
+                                                      AElementOp,
+                                                      Multiply,
+                                                      Add>>>& instances)
+{
+    add_device_operation_instances(instances,
+                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_comp_instances<
+                                       ck::Tuple<B0Layout, B1Layout>,
+                                       ck::Tuple<D0Layout>,
+                                       ck::Tuple<B0DataType, B1DataType>,
+                                       ck::Tuple<D0DataType>,
+                                       Multiply,
+                                       Add,
+                                       GemmMNKPadding,
+                                       Interwave>{});
+    add_device_operation_instances(instances,
+                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_mem_instances<
+                                       ck::Tuple<B0Layout, B1Layout>,
+                                       ck::Tuple<D0Layout>,
+                                       ck::Tuple<B0DataType, B1DataType>,
+                                       ck::Tuple<D0DataType>,
+                                       Multiply,
+                                       Add,
+                                       GemmMNKPadding,
+                                       Interwave>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_bias_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_bias_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
index 98546de040..7cbf55e5f9 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_bias_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_bias_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved.
 
 #include <cstdlib>
 
@@ -52,112 +52,6 @@ void add_device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_bias_gelu_v1_instances(
                                        Interwave>{});
 }
 
-void add_device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_bias_v1_instances(
-    std::vector<std::unique_ptr<DeviceGemmMultipleABD<AsLayout,
-                                                      ck::Tuple<B0Layout, B1Layout>,
-                                                      ck::Tuple<D0Layout>,
-                                                      ELayout,
-                                                      AsDataType,
-                                                      ck::Tuple<B0DataType, B1DataType>,
-                                                      ck::Tuple<D0DataType>,
-                                                      EDataType,
-                                                      AElementOp,
-                                                      Multiply,
-                                                      Add>>>& instances)
-{
-    add_device_operation_instances(instances,
-                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_comp_instances<
-                                       ck::Tuple<B0Layout, B1Layout>,
-                                       ck::Tuple<D0Layout>,
-                                       ck::Tuple<B0DataType, B1DataType>,
-                                       ck::Tuple<D0DataType>,
-                                       Multiply,
-                                       Add,
-                                       GemmMNKPadding,
-                                       Interwave>{});
-    add_device_operation_instances(instances,
-                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_mem_instances<
-                                       ck::Tuple<B0Layout, B1Layout>,
-                                       ck::Tuple<D0Layout>,
-                                       ck::Tuple<B0DataType, B1DataType>,
-                                       ck::Tuple<D0DataType>,
-                                       Multiply,
-                                       Add,
-                                       GemmMNKPadding,
-                                       Interwave>{});
-}
-
-void add_device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_v1_instances(
-    std::vector<std::unique_ptr<DeviceGemmMultipleABD<AsLayout,
-                                                      ck::Tuple<B0Layout, B1Layout>,
-                                                      ck::Tuple<>,
-                                                      ELayout,
-                                                      AsDataType,
-                                                      ck::Tuple<B0DataType, B1DataType>,
-                                                      ck::Tuple<>,
-                                                      EDataType,
-                                                      AElementOp,
-                                                      Multiply,
-                                                      PassThrough>>>& instances)
-{
-    add_device_operation_instances(instances,
-                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_comp_instances<
-                                       ck::Tuple<B0Layout, B1Layout>,
-                                       ck::Tuple<>,
-                                       ck::Tuple<B0DataType, B1DataType>,
-                                       ck::Tuple<>,
-                                       Multiply,
-                                       PassThrough,
-                                       GemmMNKPadding,
-                                       Interwave>{});
-    add_device_operation_instances(instances,
-                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_mem_instances<
-                                       ck::Tuple<B0Layout, B1Layout>,
-                                       ck::Tuple<>,
-                                       ck::Tuple<B0DataType, B1DataType>,
-                                       ck::Tuple<>,
-                                       Multiply,
-                                       PassThrough,
-                                       GemmMNKPadding,
-                                       Interwave>{});
-}
-
-void add_device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_gelu_v1_instances(
-    std::vector<std::unique_ptr<DeviceGemmMultipleABD<AsLayout,
-                                                      ck::Tuple<B0Layout, B1Layout>,
-                                                      ck::Tuple<>,
-                                                      ELayout,
-                                                      AsDataType,
-                                                      ck::Tuple<B0DataType, B1DataType>,
-                                                      ck::Tuple<>,
-                                                      EDataType,
-                                                      AElementOp,
-                                                      Multiply,
-                                                      FastGelu>>>& instances)
-{
-    add_device_operation_instances(instances,
-                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_comp_instances<
-                                       ck::Tuple<B0Layout, B1Layout>,
-                                       ck::Tuple<>,
-                                       ck::Tuple<B0DataType, B1DataType>,
-                                       ck::Tuple<>,
-                                       Multiply,
-                                       FastGelu,
-                                       GemmMNKPadding,
-                                       Interwave>{});
-
-    add_device_operation_instances(instances,
-                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_mem_instances<
-                                       ck::Tuple<B0Layout, B1Layout>,
-                                       ck::Tuple<>,
-                                       ck::Tuple<B0DataType, B1DataType>,
-                                       ck::Tuple<>,
-                                       Multiply,
-                                       FastGelu,
-                                       GemmMNKPadding,
-                                       Interwave>{});
-}
-
 } // namespace instance
 } // namespace device
 } // namespace tensor_operation
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
new file mode 100644
index 0000000000..044fc26721
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
@@ -0,0 +1,59 @@
+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved.
+
+#include <cstdlib>
+
+#include "ck/ck.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_abd_xdl_cshuffle.hpp"
+
+#include "device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_common.hpp"
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+void add_device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_gelu_v1_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleABD<AsLayout,
+                                                      ck::Tuple<B0Layout, B1Layout>,
+                                                      ck::Tuple<>,
+                                                      ELayout,
+                                                      AsDataType,
+                                                      ck::Tuple<B0DataType, B1DataType>,
+                                                      ck::Tuple<>,
+                                                      EDataType,
+                                                      AElementOp,
+                                                      Multiply,
+                                                      FastGelu>>>& instances)
+{
+    add_device_operation_instances(instances,
+                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_comp_instances<
+                                       ck::Tuple<B0Layout, B1Layout>,
+                                       ck::Tuple<>,
+                                       ck::Tuple<B0DataType, B1DataType>,
+                                       ck::Tuple<>,
+                                       Multiply,
+                                       FastGelu,
+                                       GemmMNKPadding,
+                                       Interwave>{});
+
+    add_device_operation_instances(instances,
+                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_mem_instances<
+                                       ck::Tuple<B0Layout, B1Layout>,
+                                       ck::Tuple<>,
+                                       ck::Tuple<B0DataType, B1DataType>,
+                                       ck::Tuple<>,
+                                       Multiply,
+                                       FastGelu,
+                                       GemmMNKPadding,
+                                       Interwave>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_multiply_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_multiply_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
new file mode 100644
index 0000000000..12bcf19252
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_multiply_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
@@ -0,0 +1,58 @@
+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved.
+
+#include <cstdlib>
+
+#include "ck/ck.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_abd_xdl_cshuffle.hpp"
+
+#include "device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_common.hpp"
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+void add_device_gemm_xdl_multi_abd_multiply_bf16_i8_bf16_mk_kn_mn_v1_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleABD<AsLayout,
+                                                      ck::Tuple<B0Layout>,
+                                                      ck::Tuple<B1Layout>,
+                                                      ELayout,
+                                                      AsDataType,
+                                                      ck::Tuple<B0DataType>,
+                                                      ck::Tuple<B1DataType>,
+                                                      EDataType,
+                                                      AElementOp,
+                                                      PassThrough,
+                                                      Multiply>>>& instances)
+{
+    add_device_operation_instances(
+        instances,
+        device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_comp_instances<ck::Tuple<B0Layout>,
+                                                                       ck::Tuple<B1Layout>,
+                                                                       ck::Tuple<B0DataType>,
+                                                                       ck::Tuple<B1DataType>,
+                                                                       PassThrough,
+                                                                       Multiply,
+                                                                       GemmMNKPadding,
+                                                                       Interwave>{});
+    add_device_operation_instances(
+        instances,
+        device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_mem_instances<ck::Tuple<B0Layout>,
+                                                                      ck::Tuple<B1Layout>,
+                                                                      ck::Tuple<B0DataType>,
+                                                                      ck::Tuple<B1DataType>,
+                                                                      PassThrough,
+                                                                      Multiply,
+                                                                      GemmMNKPadding,
+                                                                      Interwave>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_multiply_bias_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_multiply_bias_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
new file mode 100644
index 0000000000..e4a04d48b3
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_multiply_bias_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
@@ -0,0 +1,58 @@
+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved.
+
+#include <cstdlib>
+
+#include "ck/ck.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_abd_xdl_cshuffle.hpp"
+
+#include "device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_common.hpp"
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+void add_device_gemm_xdl_multi_abd_multiply_bf16_i8_bf16_mk_kn_mn_bias_v1_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleABD<AsLayout,
+                                                      ck::Tuple<B0Layout>,
+                                                      ck::Tuple<D0Layout, B1Layout>,
+                                                      ELayout,
+                                                      AsDataType,
+                                                      ck::Tuple<B0DataType>,
+                                                      ck::Tuple<D0DataType, B1DataType>,
+                                                      EDataType,
+                                                      AElementOp,
+                                                      PassThrough,
+                                                      MultiplyAdd>>>& instances)
+{
+    add_device_operation_instances(instances,
+                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_comp_instances<
+                                       ck::Tuple<B0Layout>,
+                                       ck::Tuple<D0Layout, B1Layout>,
+                                       ck::Tuple<B0DataType>,
+                                       ck::Tuple<D0DataType, B1DataType>,
+                                       PassThrough,
+                                       MultiplyAdd,
+                                       GemmMNKPadding,
+                                       Interwave>{});
+    add_device_operation_instances(instances,
+                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_mem_instances<
+                                       ck::Tuple<B0Layout>,
+                                       ck::Tuple<D0Layout, B1Layout>,
+                                       ck::Tuple<B0DataType>,
+                                       ck::Tuple<D0DataType, B1DataType>,
+                                       PassThrough,
+                                       MultiplyAdd,
+                                       GemmMNKPadding,
+                                       Interwave>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_multiply_bias_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_multiply_bias_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
index 5c46730ea7..590e89284f 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_multiply_bias_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_multiply_bias_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved.
 
 #include <cstdlib>
 
@@ -52,111 +52,6 @@ void add_device_gemm_xdl_multi_abd_multiply_bf16_i8_bf16_mk_kn_mn_bias_gelu_v1_i
                                        Interwave>{});
 }
 
-void add_device_gemm_xdl_multi_abd_multiply_bf16_i8_bf16_mk_kn_mn_bias_v1_instances(
-    std::vector<std::unique_ptr<DeviceGemmMultipleABD<AsLayout,
-                                                      ck::Tuple<B0Layout>,
-                                                      ck::Tuple<D0Layout, B1Layout>,
-                                                      ELayout,
-                                                      AsDataType,
-                                                      ck::Tuple<B0DataType>,
-                                                      ck::Tuple<D0DataType, B1DataType>,
-                                                      EDataType,
-                                                      AElementOp,
-                                                      PassThrough,
-                                                      MultiplyAdd>>>& instances)
-{
-    add_device_operation_instances(instances,
-                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_comp_instances<
-                                       ck::Tuple<B0Layout>,
-                                       ck::Tuple<D0Layout, B1Layout>,
-                                       ck::Tuple<B0DataType>,
-                                       ck::Tuple<D0DataType, B1DataType>,
-                                       PassThrough,
-                                       MultiplyAdd,
-                                       GemmMNKPadding,
-                                       Interwave>{});
-    add_device_operation_instances(instances,
-                                   device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_mem_instances<
-                                       ck::Tuple<B0Layout>,
-                                       ck::Tuple<D0Layout, B1Layout>,
-                                       ck::Tuple<B0DataType>,
-                                       ck::Tuple<D0DataType, B1DataType>,
-                                       PassThrough,
-                                       MultiplyAdd,
-                                       GemmMNKPadding,
-                                       Interwave>{});
-}
-
-void add_device_gemm_xdl_multi_abd_multiply_bf16_i8_bf16_mk_kn_mn_v1_instances(
-    std::vector<std::unique_ptr<DeviceGemmMultipleABD<AsLayout,
-                                                      ck::Tuple<B0Layout>,
-                                                      ck::Tuple<B1Layout>,
-                                                      ELayout,
-                                                      AsDataType,
-                                                      ck::Tuple<B0DataType>,
-                                                      ck::Tuple<B1DataType>,
-                                                      EDataType,
-                                                      AElementOp,
-                                                      PassThrough,
-                                                      Multiply>>>& instances)
-{
-    add_device_operation_instances(
-        instances,
-        device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_comp_instances<ck::Tuple<B0Layout>,
-                                                                       ck::Tuple<B1Layout>,
-                                                                       ck::Tuple<B0DataType>,
-                                                                       ck::Tuple<B1DataType>,
-                                                                       PassThrough,
-                                                                       Multiply,
-                                                                       GemmMNKPadding,
-                                                                       Interwave>{});
-    add_device_operation_instances(
-        instances,
-        device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_mem_instances<ck::Tuple<B0Layout>,
-                                                                      ck::Tuple<B1Layout>,
-                                                                      ck::Tuple<B0DataType>,
-                                                                      ck::Tuple<B1DataType>,
-                                                                      PassThrough,
-                                                                      Multiply,
-                                                                      GemmMNKPadding,
-                                                                      Interwave>{});
-}
-
-void add_device_gemm_xdl_multi_abd_multiply_bf16_i8_bf16_mk_kn_mn_gelu_v1_instances(
-    std::vector<std::unique_ptr<DeviceGemmMultipleABD<AsLayout,
-                                                      ck::Tuple<B0Layout>,
-                                                      ck::Tuple<B1Layout>,
-                                                      ELayout,
-                                                      AsDataType,
-                                                      ck::Tuple<B0DataType>,
-                                                      ck::Tuple<B1DataType>,
-                                                      EDataType,
-                                                      AElementOp,
-                                                      PassThrough,
-                                                      MultiplyFastGelu>>>& instances)
-{
-    add_device_operation_instances(
-        instances,
-        device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_comp_instances<ck::Tuple<B0Layout>,
-                                                                       ck::Tuple<B1Layout>,
-                                                                       ck::Tuple<B0DataType>,
-                                                                       ck::Tuple<B1DataType>,
-                                                                       PassThrough,
-                                                                       MultiplyFastGelu,
-                                                                       GemmMNKPadding,
-                                                                       Interwave>{});
-    add_device_operation_instances(
-        instances,
-        device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_mem_instances<ck::Tuple<B0Layout>,
-                                                                      ck::Tuple<B1Layout>,
-                                                                      ck::Tuple<B0DataType>,
-                                                                      ck::Tuple<B1DataType>,
-                                                                      PassThrough,
-                                                                      MultiplyFastGelu,
-                                                                      GemmMNKPadding,
-                                                                      Interwave>{});
-}
-
 } // namespace instance
 } // namespace device
 } // namespace tensor_operation
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_multiply_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_multiply_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
new file mode 100644
index 0000000000..5741ee29ac
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/gemm_multi_abd/device_gemm_xdl_multi_abd_multiply_gelu_bf16_i8_bf16_mk_kn_mn_v1_instance.cpp
@@ -0,0 +1,58 @@
+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved.
+
+#include <cstdlib>
+
+#include "ck/ck.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_abd_xdl_cshuffle.hpp"
+
+#include "device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_common.hpp"
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+void add_device_gemm_xdl_multi_abd_multiply_bf16_i8_bf16_mk_kn_mn_gelu_v1_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleABD<AsLayout,
+                                                      ck::Tuple<B0Layout>,
+                                                      ck::Tuple<B1Layout>,
+                                                      ELayout,
+                                                      AsDataType,
+                                                      ck::Tuple<B0DataType>,
+                                                      ck::Tuple<B1DataType>,
+                                                      EDataType,
+                                                      AElementOp,
+                                                      PassThrough,
+                                                      MultiplyFastGelu>>>& instances)
+{
+    add_device_operation_instances(
+        instances,
+        device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_comp_instances<ck::Tuple<B0Layout>,
+                                                                       ck::Tuple<B1Layout>,
+                                                                       ck::Tuple<B0DataType>,
+                                                                       ck::Tuple<B1DataType>,
+                                                                       PassThrough,
+                                                                       MultiplyFastGelu,
+                                                                       GemmMNKPadding,
+                                                                       Interwave>{});
+    add_device_operation_instances(
+        instances,
+        device_gemm_xdl_multi_abd_bf16_i8_bf16_mk_kn_mn_mem_instances<ck::Tuple<B0Layout>,
+                                                                      ck::Tuple<B1Layout>,
+                                                                      ck::Tuple<B0DataType>,
+                                                                      ck::Tuple<B1DataType>,
+                                                                      PassThrough,
+                                                                      MultiplyFastGelu,
+                                                                      GemmMNKPadding,
+                                                                      Interwave>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck