diff --git a/example/69_gemm_add_relu/CMakeLists.txt b/example/69_gemm_add_relu/CMakeLists.txt
index 936e9acea3..fe9d783755 100644
--- a/example/69_gemm_add_relu/CMakeLists.txt
+++ b/example/69_gemm_add_relu/CMakeLists.txt
@@ -1,9 +1,7 @@
 add_custom_target(example_gemm_add_relu_xdl)
 
-add_library(example_gemm_add_relu_xdl_fp16 gemm_add_relu_xdl_fp16.cpp)
 add_example_executable(example_gemm_add_relu_xdl_fp16 gemm_add_relu_xdl_fp16.cpp)
 
-add_library(example_gemm_add_relu_xdl_bf16 gemm_add_relu_xdl_bf16.cpp)
 add_example_executable(example_gemm_add_relu_xdl_bf16 gemm_add_relu_xdl_bf16.cpp)
 
 
@@ -12,8 +10,6 @@ add_example_executable(example_gemm_add_relu_wmma_bf16 gemm_add_relu_wmma_bf16.c
 
 add_example_executable(example_gemm_add_relu_wmma_fp16 gemm_add_relu_wmma_fp16.cpp)
 
-add_example_executable(example_gemm_add_relu_wmma_v3_fp16 gemm_add_relu_wmma_v3_fp16.cpp)
-add_example_executable(example_gemm_add_relu_wmma_v3_bf16 gemm_add_relu_wmma_v3_bf16.cpp)
 
 
 
diff --git a/example/69_gemm_add_relu/gemm_add_relu_wmma_v3_bf16.cpp b/example/69_gemm_add_relu/gemm_add_relu_wmma_bf16.cpp
similarity index 97%
rename from example/69_gemm_add_relu/gemm_add_relu_wmma_v3_bf16.cpp
rename to example/69_gemm_add_relu/gemm_add_relu_wmma_bf16.cpp
index 85b2a65b20..c91f2220aa 100644
--- a/example/69_gemm_add_relu/gemm_add_relu_wmma_v3_bf16.cpp
+++ b/example/69_gemm_add_relu/gemm_add_relu_wmma_bf16.cpp
@@ -73,6 +73,6 @@ using DeviceOpInstance = ck::tensor_operation::device::DeviceGemmMultipleD_Wmma_
 
 // clang-format on
 
-#include "run_gemm_add_relu_example_v3.inc"
+#include "run_gemm_add_relu_example.inc"
 
 int main(int argc, char* argv[]) { return !run_gemm_add_relu_example(argc, argv); }
diff --git a/example/69_gemm_add_relu/gemm_add_relu_wmma_v3_fp16.cpp b/example/69_gemm_add_relu/gemm_add_relu_wmma_fp16.cpp
similarity index 96%
rename from example/69_gemm_add_relu/gemm_add_relu_wmma_v3_fp16.cpp
rename to example/69_gemm_add_relu/gemm_add_relu_wmma_fp16.cpp
index 1a172b20a6..1d7febe66e 100644
--- a/example/69_gemm_add_relu/gemm_add_relu_wmma_v3_fp16.cpp
+++ b/example/69_gemm_add_relu/gemm_add_relu_wmma_fp16.cpp
@@ -71,6 +71,6 @@ using DeviceOpInstance = ck::tensor_operation::device::DeviceGemmMultipleD_Wmma_
 
 // clang-format on
 
-#include "run_gemm_add_relu_example_v3.inc"
+#include "run_gemm_add_relu_example.inc"
 
 int main(int argc, char* argv[]) { return !run_gemm_add_relu_example(argc, argv); }
diff --git a/example/69_gemm_add_relu/run_gemm_add_relu_example_v3.inc b/example/69_gemm_add_relu/run_gemm_add_relu_example.inc
similarity index 100%
rename from example/69_gemm_add_relu/run_gemm_add_relu_example_v3.inc
rename to example/69_gemm_add_relu/run_gemm_add_relu_example.inc
diff --git a/library/include/ck/library/tensor_operation_instance/gpu/gemm_add_relu.hpp b/library/include/ck/library/tensor_operation_instance/gpu/gemm_add_relu.hpp
index 2cc7cab5e6..0792e3eb89 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/gemm_add_relu.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/gemm_add_relu.hpp
@@ -45,30 +45,30 @@ void add_device_gemm_add_relu_xdl_c_shuffle_bf16_i8_bf16_bf16_mk_kn_mn_mn_instan
 
 #elif defined(CK_USE_WMMA)
 void add_device_gemm_add_relu_wmma_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instances(
-    std::vector<std::unique_ptr<DeviceGemmMultipleD<Row,
-                                                    Row,
-                                                    Row_Tuple,
-                                                    Row,
-                                                    F16,
-                                                    F16,
-                                                    F16_Tuple,
-                                                    F16,
-                                                    PassThrough,
-                                                    PassThrough,
-                                                    AddRelu>>>&);
+    std::vector<std::unique_ptr<DeviceGemmMultipleDSplitK<Row,
+                                                          Row,
+                                                          Row_Tuple,
+                                                          Row,
+                                                          F16,
+                                                          F16,
+                                                          F16_Tuple,
+                                                          F16,
+                                                          PassThrough,
+                                                          PassThrough,
+                                                          AddRelu>>>&);
 
 void add_device_gemm_add_relu_wmma_c_shuffle_bf16_bf16_bf16_bf16_mk_kn_mn_mn_instances(
-    std::vector<std::unique_ptr<DeviceGemmMultipleD<Row,
-                                                    Row,
-                                                    Row_Tuple,
-                                                    Row,
-                                                    BF16,
-                                                    BF16,
-                                                    BF16_Tuple,
-                                                    BF16,
-                                                    PassThrough,
-                                                    PassThrough,
-                                                    AddRelu>>>&);
+    std::vector<std::unique_ptr<DeviceGemmMultipleDSplitK<Row,
+                                                          Row,
+                                                          Row_Tuple,
+                                                          Row,
+                                                          BF16,
+                                                          BF16,
+                                                          BF16_Tuple,
+                                                          BF16,
+                                                          PassThrough,
+                                                          PassThrough,
+                                                          AddRelu>>>&);
 #endif
 
 // GEMM + Add + Relu
@@ -137,7 +137,7 @@ struct DeviceOperationInstanceFactory<
 #endif
 
 #elif defined(CK_USE_WMMA)
-        // For wmma ADataType must be same as BDatatype.
+
 #if defined(CK_ENABLE_FP16)
         if constexpr(is_same_v<ADataType, half_t> && is_same_v<BDataType, half_t> &&
                      is_same_v<D0DataType, half_t> && is_same_v<EDataType, half_t>)
@@ -151,7 +151,6 @@ struct DeviceOperationInstanceFactory<
         }
 #endif
 
-// For wmma ADataType must be same as BDatatype.
 #if defined(CK_ENABLE_BF16)
         if constexpr(is_same_v<ADataType, ck::bhalf_t> && is_same_v<BDataType, ck::bhalf_t> &&
                      is_same_v<D0DataType, ck::bhalf_t> && is_same_v<EDataType, ck::bhalf_t>)
diff --git a/library/src/tensor_operation_instance/gpu/gemm_add_relu/CMakeLists.txt b/library/src/tensor_operation_instance/gpu/gemm_add_relu/CMakeLists.txt
index 8fb7f7fb72..28e0ccb33d 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_add_relu/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/gemm_add_relu/CMakeLists.txt
@@ -4,8 +4,5 @@ add_instance_library(device_gemm_add_relu_instance
    device_gemm_add_relu_xdl_c_shuffle_bf16_i8_bf16_bf16_mk_kn_mn_mn_instance.cpp
    device_gemm_add_relu_wmma_c_shuffle_bf16_bf16_bf16_bf16_mk_kn_mn_mn_instance.cpp
    device_gemm_add_relu_wmma_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instance.cpp
-   device_gemm_add_relu_wmma_c_shuffle_v3_bf16_bf16_bf16_bf16_mk_kn_mn_mn_instance.cpp
-   device_gemm_add_relu_wmma_c_shuffle_v3_f16_f16_f16_f16_mk_kn_mn_mn_instance.cpp
 )
 
-
diff --git a/library/src/tensor_operation_instance/gpu/gemm_add_relu/device_gemm_add_relu_wmma_c_shuffle_v3_bf16_bf16_bf16_bf16_mk_kn_mn_mn_instance.cpp b/library/src/tensor_operation_instance/gpu/gemm_add_relu/device_gemm_add_relu_wmma_c_shuffle_bf16_bf16_bf16_bf16_mk_kn_mn_mn_instance.cpp
similarity index 100%
rename from library/src/tensor_operation_instance/gpu/gemm_add_relu/device_gemm_add_relu_wmma_c_shuffle_v3_bf16_bf16_bf16_bf16_mk_kn_mn_mn_instance.cpp
rename to library/src/tensor_operation_instance/gpu/gemm_add_relu/device_gemm_add_relu_wmma_c_shuffle_bf16_bf16_bf16_bf16_mk_kn_mn_mn_instance.cpp
diff --git a/library/src/tensor_operation_instance/gpu/gemm_add_relu/device_gemm_add_relu_wmma_c_shuffle_v3_f16_f16_f16_f16_mk_kn_mn_mn_instance.cpp b/library/src/tensor_operation_instance/gpu/gemm_add_relu/device_gemm_add_relu_wmma_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instance.cpp
similarity index 100%
rename from library/src/tensor_operation_instance/gpu/gemm_add_relu/device_gemm_add_relu_wmma_c_shuffle_v3_f16_f16_f16_f16_mk_kn_mn_mn_instance.cpp
rename to library/src/tensor_operation_instance/gpu/gemm_add_relu/device_gemm_add_relu_wmma_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instance.cpp