diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_bilinear_instance.hpp b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_bilinear_instance.hpp
index 75655dda8d..87b47e54f1 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_bilinear_instance.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_bilinear_instance.hpp
@@ -149,7 +149,7 @@ template <ck::index_t NDimSpatial,
           typename BLayout,
           typename ELayout,
           ConvolutionBackwardWeightSpecialization ConvSpec>
-using device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_bilinear_instances = std::tuple<
+using device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_f32_bf16_bilinear_instances = std::tuple<
     // clang-format off
         //#########################################|     Num| InLayout| WeiLayout| OutLayout| DsData| InData| WeiData| OutData| AccData| DsData|          In|         Wei|         Out|              ConvBackward| Block|  MPer|  NPer| K0Per| K1| MPer| NPer| MXdl| NXdl|  ABlockTransfer|   ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds|    CShuffle|    CShuffle|   CBlockTransfer|  CBlockTransfer|
         //#########################################|     Dim|         |          |          | Layout|   Type|    Type|    Type|    Type|   Type| Elementwise| Elementwise| Elementwise|                    Weight|  Size| Block| Block| Block|   |  XDL|  XDL|  Per|  Per|   ThreadCluster|    ThreadCluster| SrcAccessOrder|   SrcVectorDim|      SrcScalar|      DstScalar| AddExtraM|   ThreadCluster|  ThreadCluster| SrcAccessOrder|  SrcVectorDim|      SrcScalar|      DstScalar| AddExtraN| MXdlPerWave| NXdlPerWave|   ClusterLengths| ScalarPerVector|
@@ -177,6 +177,22 @@ using device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_bilinear_instances = std
     // clang-format on
     >;
 
+template <ck::index_t NDimSpatial,
+          typename ALayout,
+          typename BLayout,
+          typename ELayout,
+          ConvolutionBackwardWeightSpecialization ConvSpec>
+using device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_bilinear_instances = std::tuple<
+    // clang-format off
+        //#########################################|     Num| InLayout| WeiLayout| OutLayout| DsData| InData| WeiData| OutData| AccData| DsData|          In|         Wei|         Out|              ConvBackward| Block|  MPer|  NPer| K0Per| K1| MPer| NPer| MXdl| NXdl|  ABlockTransfer|   ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds|    CShuffle|    CShuffle|   CBlockTransfer|  CBlockTransfer|
+        //#########################################|     Dim|         |          |          | Layout|   Type|    Type|    Type|    Type|   Type| Elementwise| Elementwise| Elementwise|                    Weight|  Size| Block| Block| Block|   |  XDL|  XDL|  Per|  Per|   ThreadCluster|    ThreadCluster| SrcAccessOrder|   SrcVectorDim|      SrcScalar|      DstScalar| AddExtraM|   ThreadCluster|  ThreadCluster| SrcAccessOrder|  SrcVectorDim|      SrcScalar|      DstScalar| AddExtraN| MXdlPerWave| NXdlPerWave|   ClusterLengths| ScalarPerVector|
+        //#########################################| Spatial|         |          |          |       |       |        |        |        |       |   Operation|   Operation|   Operation|            Specialization|      |      |      |      |   |     |     | Wave| Wave| Lengths_K0_M_K1|     ArrangeOrder|               |               |      PerVector|   PerVector_K1|          | Lengths_K0_N_K1|   ArrangeOrder|               |              |      PerVector|   PerVector_K1|          |  PerShuffle|  PerShuffle| MBlock_MPerBlock|    NWaveNPerXdl|
+        //#########################################|        |         |          |          |       |       |        |        |        |       |            |            |            |                          |      |      |      |      |   |     |     |     |     |                |                 |               |               |               |               |          |                |               |               |              |               |               |          |            |            | NBlock_NPerBlock|                |
+        // generic instance
+        DeviceGroupedConvBwdWeightMultipleD_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,    ELayout, Tuple<BLayout>, BF16,     BF16,    BF16,     F32, Tuple<BF16>, PassThrough, Bilinear, PassThrough,   ConvSpec,    64,    64,    64,     4,  8,   32,   32,    2,    2,  S<1, 4, 8,  2>, S<0, 3, 1, 2>,   S<0, 2, 1, 3>,              2,              1,              4,      true,   S<1, 4, 8,  2>,   S<0, 3, 1, 2>,  S<0, 2, 1, 3>,             2,              1,              4,      true,           1,           1,   S<1, 16, 1, 4>,               1>
+    // clang-format on
+    >;
+
 template <ck::index_t NDimSpatial,
           typename ALayout,
           typename BLayout,
diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_scale_instance.hpp b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_scale_instance.hpp
index ed1cb3997f..c2753ff01a 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_scale_instance.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_scale_instance.hpp
@@ -148,7 +148,7 @@ template <ck::index_t NDimSpatial,
           typename BLayout,
           typename ELayout,
           ConvolutionBackwardWeightSpecialization ConvSpec>
-using device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_scale_instances = std::tuple<
+using device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_f32_bf16_scale_instances = std::tuple<
     // clang-format off
         //#########################################|     Num| InLayout| WeiLayout| OutLayout| DsData| InData| WeiData| OutData| AccData| DsData|          In|         Wei|         Out|              ConvBackward| Block|  MPer|  NPer| K0Per| K1| MPer| NPer| MXdl| NXdl|  ABlockTransfer|   ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds|    CShuffle|    CShuffle|   CBlockTransfer|  CBlockTransfer|
         //#########################################|     Dim|         |          |          | Layout|   Type|    Type|    Type|    Type|   Type| Elementwise| Elementwise| Elementwise|                    Weight|  Size| Block| Block| Block|   |  XDL|  XDL|  Per|  Per|   ThreadCluster|    ThreadCluster| SrcAccessOrder|   SrcVectorDim|      SrcScalar|      DstScalar| AddExtraM|   ThreadCluster|  ThreadCluster| SrcAccessOrder|  SrcVectorDim|      SrcScalar|      DstScalar| AddExtraN| MXdlPerWave| NXdlPerWave|   ClusterLengths| ScalarPerVector|
@@ -176,6 +176,22 @@ using device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_scale_instances = std::t
     // clang-format on
     >;
 
+template <ck::index_t NDimSpatial,
+          typename ALayout,
+          typename BLayout,
+          typename ELayout,
+          ConvolutionBackwardWeightSpecialization ConvSpec>
+using device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_scale_instances = std::tuple<
+    // clang-format off
+        //#########################################|     Num| InLayout| WeiLayout| OutLayout| DsData| InData| WeiData| OutData| AccData| DsData|          In|         Wei|         Out|              ConvBackward| Block|  MPer|  NPer| K0Per| K1| MPer| NPer| MXdl| NXdl|  ABlockTransfer|   ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds|    CShuffle|    CShuffle|   CBlockTransfer|  CBlockTransfer|
+        //#########################################|     Dim|         |          |          | Layout|   Type|    Type|    Type|    Type|   Type| Elementwise| Elementwise| Elementwise|                    Weight|  Size| Block| Block| Block|   |  XDL|  XDL|  Per|  Per|   ThreadCluster|    ThreadCluster| SrcAccessOrder|   SrcVectorDim|      SrcScalar|      DstScalar| AddExtraM|   ThreadCluster|  ThreadCluster| SrcAccessOrder|  SrcVectorDim|      SrcScalar|      DstScalar| AddExtraN| MXdlPerWave| NXdlPerWave|   ClusterLengths| ScalarPerVector|
+        //#########################################| Spatial|         |          |          |       |       |        |        |        |       |   Operation|   Operation|   Operation|            Specialization|      |      |      |      |   |     |     | Wave| Wave| Lengths_K0_M_K1|     ArrangeOrder|               |               |      PerVector|   PerVector_K1|          | Lengths_K0_N_K1|   ArrangeOrder|               |              |      PerVector|   PerVector_K1|          |  PerShuffle|  PerShuffle| MBlock_MPerBlock|    NWaveNPerXdl|
+        //#########################################|        |         |          |          |       |       |        |        |        |       |            |            |            |                          |      |      |      |      |   |     |     |     |     |                |                 |               |               |               |               |          |                |               |               |              |               |               |          |            |            | NBlock_NPerBlock|                |
+        // generic instance
+        DeviceGroupedConvBwdWeightMultipleD_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,    ELayout, Empty_Tuple, BF16,     BF16,    BF16,     F32, Empty_Tuple, PassThrough, Scale, PassThrough,       ConvSpec,    64,    64,    64,     4,  8,   32,   32,    2,    2,  S<1, 4, 8,  2>, S<0, 3, 1, 2>,   S<0, 2, 1, 3>,              2,              1,              4,      true,   S<1, 4, 8,  2>,   S<0, 3, 1, 2>,  S<0, 2, 1, 3>,             2,              1,              4,      true,           1,           1,   S<1, 16, 1, 4>,               1>
+    // clang-format on
+    >;
+
 template <ck::index_t NDimSpatial,
           typename ALayout,
           typename BLayout,
diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_backward_weight_bilinear.hpp b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_backward_weight_bilinear.hpp
index 6fde57b44e..846ba935e5 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_backward_weight_bilinear.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_backward_weight_bilinear.hpp
@@ -11,6 +11,7 @@
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 
 #include "ck/library/tensor_operation_instance/device_operation_instance_factory.hpp"
+#include "ck/utility/data_type.hpp"
 
 namespace ck {
 namespace tensor_operation {
@@ -65,6 +66,19 @@ void add_device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16
                                                                     PassThrough,
                                                                     Bilinear,
                                                                     PassThrough>>>& instances);
+void add_device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_instances(
+    std::vector<std::unique_ptr<DeviceGroupedConvBwdWeightMultipleD<3,
+                                                                    NDHWGC,
+                                                                    GKZYXC,
+                                                                    NDHWGK,
+                                                                    Tuple<GKZYXC>,
+                                                                    BF16,
+                                                                    BF16,
+                                                                    BF16,
+                                                                    Tuple<BF16>,
+                                                                    PassThrough,
+                                                                    Bilinear,
+                                                                    PassThrough>>>& instances);
 #endif
 #ifdef CK_ENABLE_FP16
 void add_device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_f16_instances(
@@ -255,6 +269,16 @@ struct DeviceOperationInstanceFactory<
                     add_device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_f32_bf16_instances(
                         op_ptrs);
                 }
+
+                if constexpr(is_same_v<InDataType, ck::bhalf_t> &&
+                             is_same_v<WeiDataType, ck::bhalf_t> &&
+                             is_same_v<OutDataType, ck::bhalf_t> &&
+                             is_same_v<ComputeTypeA, ck::bhalf_t> &&
+                             is_same_v<ComputeTypeB, ck::bhalf_t>)
+                {
+                    add_device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_instances(
+                        op_ptrs);
+                }
 #endif
 #if defined CK_ENABLE_FP16 && defined CK_ENABLE_FP8 && defined CK_ENABLE_BF8
                 if constexpr(is_same_v<InDataType, half_t> && is_same_v<WeiDataType, half_t> &&
diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_backward_weight_scale.hpp b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_backward_weight_scale.hpp
index c24243943a..0ad9c38714 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_backward_weight_scale.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_backward_weight_scale.hpp
@@ -66,6 +66,19 @@ void add_device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_f3
                                                                     PassThrough,
                                                                     Scale,
                                                                     PassThrough>>>& instances);
+void add_device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_instances(
+    std::vector<std::unique_ptr<DeviceGroupedConvBwdWeightMultipleD<3,
+                                                                    NDHWGC,
+                                                                    GKZYXC,
+                                                                    NDHWGK,
+                                                                    Tuple<>,
+                                                                    BF16,
+                                                                    BF16,
+                                                                    BF16,
+                                                                    Tuple<>,
+                                                                    PassThrough,
+                                                                    Scale,
+                                                                    PassThrough>>>& instances);
 #endif
 #ifdef CK_ENABLE_FP16
 void add_device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_f16_instances(
@@ -255,6 +268,16 @@ struct DeviceOperationInstanceFactory<
                     add_device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_f32_bf16_instances(
                         op_ptrs);
                 }
+                if constexpr(is_same_v<InDataType, ck::bhalf_t> &&
+                             is_same_v<WeiDataType, ck::bhalf_t> &&
+                             is_same_v<OutDataType, ck::bhalf_t> &&
+                             is_same_v<ComputeTypeA, ck::bhalf_t> &&
+                             is_same_v<ComputeTypeB, ck::bhalf_t>)
+                {
+                    add_device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_instances(
+                        op_ptrs);
+                }
+
 #endif
 #if defined CK_ENABLE_FP16 && defined CK_ENABLE_FP8 && defined CK_ENABLE_BF8
                 if constexpr(is_same_v<InDataType, half_t> && is_same_v<WeiDataType, half_t> &&
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_bilinear/CMakeLists.txt b/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_bilinear/CMakeLists.txt
index 455f14d9f4..9cb0ccda0d 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_bilinear/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_bilinear/CMakeLists.txt
@@ -6,7 +6,9 @@ set(GROUPED_CONV3D_BWD_WEIGHT_BILINEAR
     xdl/device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_f16_instance.cpp
     xdl/device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_f32_instance.cpp
     xdl/device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_f32_tf32_instance.cpp
-    xdl/device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp)
+    xdl/device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp
+    xdl/device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_f32_bf16_instance.cpp
+    )
 
 if((DTYPES MATCHES "fp8" AND DTYPES MATCHES "bf8" AND DTYPES MATCHES "fp16") OR NOT DEFINED DTYPES)
     list(APPEND GROUPED_CONV3D_BWD_WEIGHT_BILINEAR
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_bilinear/xdl/device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_f32_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_bilinear/xdl/device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_f32_bf16_instance.cpp
new file mode 100644
index 0000000000..85d972d64a
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_bilinear/xdl/device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_f32_bf16_instance.cpp
@@ -0,0 +1,50 @@
+// Copyright (c) Advanced Micro Devices, Inc., or its affiliates.
+// SPDX-License-Identifier: MIT
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_bilinear_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
+void add_device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_f32_bf16_instances(
+    std::vector<std::unique_ptr<DeviceGroupedConvBwdWeightMultipleD<3,
+                                                                    NDHWGC,
+                                                                    GKZYXC,
+                                                                    NDHWGK,
+                                                                    Tuple<GKZYXC>,
+                                                                    BF16,
+                                                                    F32,
+                                                                    BF16,
+                                                                    Tuple<F32>,
+                                                                    PassThrough,
+                                                                    Bilinear,
+                                                                    PassThrough>>>& instances)
+{
+    // 1. Default
+    add_device_operation_instances(
+        instances,
+        device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_f32_bf16_bilinear_instances<
+            3,
+            NDHWGC,
+            GKZYXC,
+            NDHWGK,
+            ConvBwdWeightDefault>{});
+    // 2. Filter1x1Stride1Pad0
+    add_device_operation_instances(
+        instances,
+        device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_f32_bf16_bilinear_instances<
+            3,
+            NDHWGC,
+            GKZYXC,
+            NDHWGK,
+            ConvBwdWeightFilter1x1Stride1Pad0>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_bilinear/xdl/device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_bilinear/xdl/device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp
index f0582697b9..1b019a8cf7 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_bilinear/xdl/device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_bilinear/xdl/device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp
@@ -9,22 +9,21 @@ namespace tensor_operation {
 namespace device {
 namespace instance {
 
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
-void add_device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_f32_bf16_instances(
+void add_device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvBwdWeightMultipleD<3,
                                                                     NDHWGC,
                                                                     GKZYXC,
                                                                     NDHWGK,
                                                                     Tuple<GKZYXC>,
                                                                     BF16,
-                                                                    F32,
                                                                     BF16,
-                                                                    Tuple<F32>,
+                                                                    BF16,
+                                                                    Tuple<BF16>,
                                                                     PassThrough,
                                                                     Bilinear,
                                                                     PassThrough>>>& instances)
 {
-    // 1. Default
+    // Default bwd weight bilinear
     add_device_operation_instances(
         instances,
         device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_bilinear_instances<
@@ -33,15 +32,6 @@ void add_device_grouped_conv3d_bwd_weight_xdl_bilinear_ndhwgc_gkzyxc_ndhwgk_bf16
             GKZYXC,
             NDHWGK,
             ConvBwdWeightDefault>{});
-    // 2. Filter1x1Stride1Pad0
-    add_device_operation_instances(
-        instances,
-        device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_bilinear_instances<
-            3,
-            NDHWGC,
-            GKZYXC,
-            NDHWGK,
-            ConvBwdWeightFilter1x1Stride1Pad0>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_scale/CMakeLists.txt b/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_scale/CMakeLists.txt
index b7fefdc94f..85c99dc422 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_scale/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_scale/CMakeLists.txt
@@ -6,7 +6,9 @@ set(GROUPED_CONV3D_BWD_WEIGHT_SCALE
     xdl/device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_f16_instance.cpp
     xdl/device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_f32_instance.cpp
     xdl/device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_f32_tf32_instance.cpp
-    xdl/device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp)
+    xdl/device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp
+    xdl/device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_f32_bf16_instance.cpp
+    )
 
 if((DTYPES MATCHES "fp8" AND DTYPES MATCHES "bf8" AND DTYPES MATCHES "fp16") OR NOT DEFINED DTYPES)
     list(APPEND GROUPED_CONV3D_BWD_WEIGHT_SCALE
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_scale/xdl/device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_f32_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_scale/xdl/device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_f32_bf16_instance.cpp
new file mode 100644
index 0000000000..41f051479b
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_scale/xdl/device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_f32_bf16_instance.cpp
@@ -0,0 +1,51 @@
+// Copyright (c) Advanced Micro Devices, Inc., or its affiliates.
+// SPDX-License-Identifier: MIT
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/device_operation_instance_factory.hpp"
+#include "ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_scale_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
+void add_device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_f32_bf16_instances(
+    std::vector<std::unique_ptr<DeviceGroupedConvBwdWeightMultipleD<3,
+                                                                    NDHWGC,
+                                                                    GKZYXC,
+                                                                    NDHWGK,
+                                                                    Tuple<>,
+                                                                    BF16,
+                                                                    F32,
+                                                                    BF16,
+                                                                    Tuple<>,
+                                                                    PassThrough,
+                                                                    Scale,
+                                                                    PassThrough>>>& instances)
+{
+    // 1. Default
+    add_device_operation_instances(
+        instances,
+        device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_f32_bf16_scale_instances<
+            3,
+            NDHWGC,
+            GKZYXC,
+            NDHWGK,
+            ConvBwdWeightDefault>{});
+    // 2. Filter1x1Stride1Pad0
+    add_device_operation_instances(
+        instances,
+        device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_f32_bf16_scale_instances<
+            3,
+            NDHWGC,
+            GKZYXC,
+            NDHWGK,
+            ConvBwdWeightFilter1x1Stride1Pad0>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_scale/xdl/device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_scale/xdl/device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp
index c1bc8c3a1c..a08c4f4054 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_scale/xdl/device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv3d_bwd_weight_scale/xdl/device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp
@@ -10,21 +10,21 @@ namespace device {
 namespace instance {
 
 // Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
-void add_device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_f32_bf16_instances(
+void add_device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvBwdWeightMultipleD<3,
                                                                     NDHWGC,
                                                                     GKZYXC,
                                                                     NDHWGK,
                                                                     Tuple<>,
                                                                     BF16,
-                                                                    F32,
+                                                                    BF16,
                                                                     BF16,
                                                                     Tuple<>,
                                                                     PassThrough,
                                                                     Scale,
                                                                     PassThrough>>>& instances)
 {
-    // 1. Default
+    // Default conv bwd weight
     add_device_operation_instances(
         instances,
         device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_scale_instances<3,
@@ -32,15 +32,6 @@ void add_device_grouped_conv3d_bwd_weight_xdl_scale_ndhwgc_gkzyxc_ndhwgk_bf16_f3
                                                                           GKZYXC,
                                                                           NDHWGK,
                                                                           ConvBwdWeightDefault>{});
-    // 2. Filter1x1Stride1Pad0
-    add_device_operation_instances(
-        instances,
-        device_grouped_conv_bwd_weight_xdl_c_shuffle_bf16_scale_instances<
-            3,
-            NDHWGC,
-            GKZYXC,
-            NDHWGK,
-            ConvBwdWeightFilter1x1Stride1Pad0>{});
 }
 
 } // namespace instance