diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_v3_xdl_instance.hpp b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_v3_xdl_instance.hpp
index 3a3dc156ec..7cc28d3850 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_v3_xdl_instance.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_v3_xdl_instance.hpp
@@ -123,7 +123,8 @@ using device_grouped_conv_bwd_weight_v3_xdl_c_shuffle_f16_direct_load_instances
         DeviceGroupedConvBwdWeight_Xdl_CShuffleV3< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                 ConvSpec,   256,    64,  128,    64,   8,     32,   32,    1,    2,     S<4,  8, 8>,      S<0, 2, 1>,S<0, 2, 1>,               1,              8,              8,          0,    S<4, 16, 4>,      S<0, 2, 1>,S<0, 2, 1>,               1,              8,              8,          0,          1,           1,                   S<1, 32, 1, 8>,               8,  Scheduler, PipelineVersion, F16,  F16, true, 2>,
         DeviceGroupedConvBwdWeight_Xdl_CShuffleV3< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                 ConvSpec,   256,    64,   64,    64,   8,     16,   16,    2,    2,     S<4,  8, 8>,      S<0, 2, 1>,S<0, 2, 1>,               1,              8,              8,          0,    S<4,  8, 8>,      S<0, 2, 1>,S<0, 2, 1>,               1,              8,              8,          0,          1,           1,                   S<1, 32, 1, 4>,               8,  Scheduler, PipelineVersion, F16,  F16, true>,
         DeviceGroupedConvBwdWeight_Xdl_CShuffleV3< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                 ConvSpec,   256,    64,   64,    64,   8,     16,   16,    2,    2,     S<4,  8, 8>,      S<0, 2, 1>,S<0, 2, 1>,               1,              8,              8,          0,    S<4,  8, 8>,      S<0, 2, 1>,S<0, 2, 1>,               1,              8,              8,          0,          1,           1,                   S<1, 32, 1, 4>,               2,  Scheduler, PipelineVersion, F16,  F16, true>,
-        DeviceGroupedConvBwdWeight_Xdl_CShuffleV3< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                 ConvSpec,   256,    64,   64,    64,   8,     16,   16,    2,    2,     S<4,  32, 2>,     S<0, 2, 1>,S<0, 2, 1>,               1,              2,              1,          0,    S<4, 32, 2>,      S<0, 2, 1>,S<0, 2, 1>,               1,              2,              1,          0,          1,           1,                   S<1, 32, 1, 4>,               4,  Scheduler, PipelineVersion, F16,  F16, true>
+        DeviceGroupedConvBwdWeight_Xdl_CShuffleV3< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                 ConvSpec,   256,    64,   64,    64,   8,     16,   16,    2,    2,     S<4,  32, 2>,     S<0, 2, 1>,S<0, 2, 1>,               1,              2,              1,          0,    S<4, 32, 2>,      S<0, 2, 1>,S<0, 2, 1>,               1,              2,              1,          0,          1,           1,                   S<1, 32, 1, 4>,               4,  Scheduler, PipelineVersion, F16,  F16, true>,
+        DeviceGroupedConvBwdWeight_Xdl_CShuffleV3< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                 ConvSpec,   256,   128,   64,    32,   8,     32,   32,    2,    1,     S<2,  16, 8>,     S<0, 2, 1>,S<0, 2, 1>,               1,              8,              8,          0,    S<2,   8, 16>,    S<0, 2, 1>,S<0, 2, 1>,               1,              8,              8,          0,          1,           1,                   S<1, 32, 1, 8>,               8,  Scheduler, PipelineVersion, F16,  F16, true>
     // clang-format on
     >;