diff --git a/include/ck/utility/dtype_vector.hpp b/include/ck/utility/dtype_vector.hpp
index 65eed0624c..4a91054fd9 100644
--- a/include/ck/utility/dtype_vector.hpp
+++ b/include/ck/utility/dtype_vector.hpp
@@ -2124,6 +2124,14 @@ using int32x16_t = typename vector_type<int32_t, 16>::type;
 using int32x32_t = typename vector_type<int32_t, 32>::type;
 using int32x64_t = typename vector_type<int32_t, 64>::type;
 
+// i16
+using int16x2_t  = typename vector_type<int16_t, 2>::type;
+using int16x4_t  = typename vector_type<int16_t, 4>::type;
+using int16x8_t  = typename vector_type<int16_t, 8>::type;
+using int16x16_t = typename vector_type<int16_t, 16>::type;
+using int16x32_t = typename vector_type<int16_t, 32>::type;
+using int16x64_t = typename vector_type<int16_t, 64>::type;
+
 // i8
 using int8x2_t  = typename vector_type<int8_t, 2>::type;
 using int8x4_t  = typename vector_type<int8_t, 4>::type;
diff --git a/library/include/ck/library/tensor_operation_instance/gpu/gemm_add.hpp b/library/include/ck/library/tensor_operation_instance/gpu/gemm_add.hpp
index 985cf2b1ec..4b50d6a351 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/gemm_add.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/gemm_add.hpp
@@ -146,15 +146,16 @@ struct DeviceOperationInstanceFactory<
 #endif
 
 #if defined(CK_ENABLE_BF16)
-// TODO:
-// here for WMMA, currently BDataType and ADataType must be the same
+        // TODO:
+        // here for WMMA, currently BDataType and ADataType must be the same
         if constexpr(is_same_v<ADataType, ck::bhalf_t> && is_same_v<BDataType, bhalf_t> &&
                      is_same_v<D0DataType, ck::bhalf_t> && is_same_v<EDataType, ck::bhalf_t>)
         {
             if constexpr(is_same_v<ALayout, Row> && is_same_v<BLayout, Row> &&
                          is_same_v<D0Layout, Row> && is_same_v<ELayout, Row>)
             {
-                add_device_gemm_add_wmma_c_shuffle_bf16_f16_bf16_bf16_mk_kn_mn_mn_instances(op_ptrs);
+                add_device_gemm_add_wmma_c_shuffle_bf16_f16_bf16_bf16_mk_kn_mn_mn_instances(
+                    op_ptrs);
             }
         }
 #endif
diff --git a/library/src/tensor_operation_instance/gpu/gemm_add/device_gemm_add_wmma_c_shuffle_bf16_f16_bf16_bf16_mk_kn_mn_mn_instance.cpp b/library/src/tensor_operation_instance/gpu/gemm_add/device_gemm_add_wmma_c_shuffle_bf16_f16_bf16_bf16_mk_kn_mn_mn_instance.cpp
index 4db8af6554..50c90e781d 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_add/device_gemm_add_wmma_c_shuffle_bf16_f16_bf16_bf16_mk_kn_mn_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_add/device_gemm_add_wmma_c_shuffle_bf16_f16_bf16_bf16_mk_kn_mn_mn_instance.cpp
@@ -60,7 +60,8 @@ void add_device_gemm_add_wmma_c_shuffle_bf16_f16_bf16_bf16_mk_kn_mn_mn_instances
                                                     Add>>>& instances)
 {
     add_device_operation_instances(
-        instances, device_gemm_add_wmma_c_shuffle_bf16_f16_bf16_bf16_mk_kn_mn_mn_generic_instances{});
+        instances,
+        device_gemm_add_wmma_c_shuffle_bf16_f16_bf16_bf16_mk_kn_mn_mn_generic_instances{});
     add_device_operation_instances(
         instances, device_gemm_add_wmma_c_shuffle_bf16_f16_bf16_bf16_mk_kn_mn_mn_instances{});
 }
diff --git a/library/src/tensor_operation_instance/gpu/gemm_add/device_gemm_add_wmma_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instance.cpp b/library/src/tensor_operation_instance/gpu/gemm_add/device_gemm_add_wmma_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instance.cpp
index 368d17002a..6dde4d8a87 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_add/device_gemm_add_wmma_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_add/device_gemm_add_wmma_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instance.cpp
@@ -43,7 +43,7 @@ using device_gemm_add_wmma_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instances = std
         DeviceGemmMultipleD_Wmma_CShuffle<     Row,    Row, Row_Tuple,    Row,   F16,    F16,      F32,      F32,F16_Tuple,   F16,  PassThrough, PassThrough,         Add, GemmMNKPadding,        1,   256,    16,   128,    32,   8,   16,   16,       1,       2,     S<4, 16, 4>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              2,              2,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              1,              8,         1,           1,           1,               S<1, 16, 1, 4>,               1, LoopScheduler::Default,        PipelineVersion::v1>,
         DeviceGemmMultipleD_Wmma_CShuffle<     Row,    Row, Row_Tuple,    Row,   F16,    F16,      F32,      F32,F16_Tuple,   F16,  PassThrough, PassThrough,         Add, GemmMNKPadding,        1,    64,    16,    16,    32,   8,   16,   16,       1,       1,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              1,              1,         1,           1,           1,               S<1, 16, 1, 4>,               1, LoopScheduler::Default,        PipelineVersion::v1>,
         DeviceGemmMultipleD_Wmma_CShuffle<     Row,    Row, Row_Tuple,    Row,   F16,    F16,      F32,      F32,F16_Tuple,   F16,  PassThrough, PassThrough,         Add, GemmMNKPadding,        1,    64,    16,    16,    64,   8,   16,   16,       1,       1,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              1,              1,         1,           1,           1,               S<1, 16, 1, 4>,               1, LoopScheduler::Default,        PipelineVersion::v1>
-        // clang-format on
+    // clang-format on
     >;
 
 void add_device_gemm_add_wmma_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instances(