diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp
index 8090b2449a..01415c2ddd 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp
@@ -304,7 +304,23 @@ struct DeviceOperationInstanceFactory<ck::tensor_operation::device::DeviceGroupe
                     op_ptrs);
             }
 #endif
-
+#ifdef CK_ENABLE_BF16
+            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<AComputeType, ck::bhalf_t> &&
+                         is_same_v<BComputeType, ck::bhalf_t>)
+            {
+                add_device_grouped_conv2d_fwd_xdl_merged_groups_ngchw_gkyxc_ngkhw_bf16_instances(
+                    op_ptrs);
+                add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_instances(op_ptrs);
+                add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_comp_instances(op_ptrs);
+                add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_mem_intra_instances(
+                    op_ptrs);
+                add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_mem_inter_instances(
+                    op_ptrs);
+            }
+#endif
 #ifdef CK_ENABLE_INT8
             if constexpr(is_same_v<InDataType, int8_t> && is_same_v<WeiDataType, int8_t> &&
                          is_same_v<OutDataType, int8_t> && is_same_v<AComputeType, int8_t> &&
diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_comp_xdl.inc b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_comp_xdl.inc
index e47a876e1b..9a83e36b99 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_comp_xdl.inc
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_comp_xdl.inc
@@ -90,6 +90,22 @@ void add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f16_comp_instances(
                                                                 PassThrough>>>& instances);
 #endif
 
+#ifdef CK_ENABLE_BF16
+void add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_comp_instances(
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
+                                                                NGCHW,
+                                                                GKYXC,
+                                                                Empty_Tuple,
+                                                                NGKHW,
+                                                                BF16,
+                                                                BF16,
+                                                                Empty_Tuple,
+                                                                BF16,
+                                                                PassThrough,
+                                                                PassThrough,
+                                                                PassThrough>>>& instances);
+#endif
+
 #ifdef CK_ENABLE_FP32
 void add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f32_comp_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_mem_inter_xdl.inc b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_mem_inter_xdl.inc
index f74622ad42..662fadadcf 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_mem_inter_xdl.inc
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_mem_inter_xdl.inc
@@ -90,6 +90,22 @@ void add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f16_mem_inter_instances
                                                                 PassThrough>>>& instances);
 #endif
 
+#ifdef CK_ENABLE_BF16
+void add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_mem_inter_instances(
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
+                                                                NGCHW,
+                                                                GKYXC,
+                                                                Empty_Tuple,
+                                                                NGKHW,
+                                                                BF16,
+                                                                BF16,
+                                                                Empty_Tuple,
+                                                                BF16,
+                                                                PassThrough,
+                                                                PassThrough,
+                                                                PassThrough>>>& instances);
+#endif
+
 #ifdef CK_ENABLE_FP32
 void add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f32_mem_inter_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_mem_intra_xdl.inc b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_mem_intra_xdl.inc
index 81737b6142..f283fe8550 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_mem_intra_xdl.inc
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_mem_intra_xdl.inc
@@ -90,6 +90,22 @@ void add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f16_mem_intra_instances
                                                                 PassThrough>>>& instances);
 #endif
 
+#ifdef CK_ENABLE_BF16
+void add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_mem_intra_instances(
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
+                                                                NGCHW,
+                                                                GKYXC,
+                                                                Empty_Tuple,
+                                                                NGKHW,
+                                                                BF16,
+                                                                BF16,
+                                                                Empty_Tuple,
+                                                                BF16,
+                                                                PassThrough,
+                                                                PassThrough,
+                                                                PassThrough>>>& instances);
+#endif
+
 #ifdef CK_ENABLE_FP32
 void add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f32_mem_intra_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_xdl.inc b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_xdl.inc
index 4cb2aae092..c977c89c94 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_xdl.inc
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_xdl.inc
@@ -204,6 +204,22 @@ void add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f16_instances(
                                                                 PassThrough>>>& instances);
 #endif
 
+#ifdef CK_ENABLE_BF16
+void add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_instances(
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
+                                                                NGCHW,
+                                                                GKYXC,
+                                                                Empty_Tuple,
+                                                                NGKHW,
+                                                                BF16,
+                                                                BF16,
+                                                                Empty_Tuple,
+                                                                BF16,
+                                                                PassThrough,
+                                                                PassThrough,
+                                                                PassThrough>>>& instances);
+#endif
+
 #ifdef CK_ENABLE_FP32
 void add_device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f32_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_xdl_merged_groups.inc b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_xdl_merged_groups.inc
index 1bd2697b97..a81e1e07ba 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_xdl_merged_groups.inc
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_xdl_merged_groups.inc
@@ -23,6 +23,20 @@ void add_device_grouped_conv2d_fwd_xdl_merged_groups_nhwgc_gkyxc_nhwgk_bf16_inst
                                                                 PassThrough,
                                                                 PassThrough,
                                                                 PassThrough>>>& instances);
+
+void add_device_grouped_conv2d_fwd_xdl_merged_groups_ngchw_gkyxc_ngkhw_bf16_instances(
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
+                                                                NGCHW,
+                                                                GKYXC,
+                                                                Empty_Tuple,
+                                                                NGKHW,
+                                                                BF16,
+                                                                BF16,
+                                                                Empty_Tuple,
+                                                                BF16,
+                                                                PassThrough,
+                                                                PassThrough,
+                                                                PassThrough>>>& instances);
 #endif
 
 #ifdef CK_ENABLE_FP16
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/CMakeLists.txt b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/CMakeLists.txt
index 98bee66a91..146916cfd3 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/CMakeLists.txt
@@ -11,6 +11,7 @@ add_instance_library(device_grouped_conv2d_fwd_instance
    xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_instance.cpp
    xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_instance.cpp
    # NGCHW, GKYXC, NGKHW
+   xdl/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_instance.cpp
    xdl/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f16_instance.cpp
    xdl/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f32_instance.cpp
    xdl/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_int8_instance.cpp
@@ -27,6 +28,7 @@ add_instance_library(device_grouped_conv2d_fwd_instance
    xdl/merged_groups/device_grouped_conv2d_fwd_xdl_merged_groups_nhwgc_gkyxc_nhwgk_f32_instance.cpp
    xdl/merged_groups/device_grouped_conv2d_fwd_xdl_merged_groups_nhwgc_gkyxc_nhwgk_int8_instance.cpp
    # NGCHW, GKYXC, NGKHW
+   xdl/merged_groups/device_grouped_conv2d_fwd_xdl_merged_groups_ngchw_gkyxc_ngkhw_bf16_instance.cpp
    xdl/merged_groups/device_grouped_conv2d_fwd_xdl_merged_groups_ngchw_gkyxc_ngkhw_f16_instance.cpp
    xdl/merged_groups/device_grouped_conv2d_fwd_xdl_merged_groups_ngchw_gkyxc_ngkhw_f32_instance.cpp
    xdl/merged_groups/device_grouped_conv2d_fwd_xdl_merged_groups_ngchw_gkyxc_ngkhw_int8_instance.cpp
@@ -42,10 +44,12 @@ add_instance_library(device_grouped_conv2d_fwd_instance
    xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_mem_inter_instance.cpp
    xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_mem_inter_instance.cpp
    # NGCHW, GKYXC, NGKHW
+   xdl/mem/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_mem_intra_instance.cpp
    xdl/mem/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f16_mem_intra_instance.cpp
    xdl/mem/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f32_mem_intra_instance.cpp
    xdl/mem/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_int8_mem_intra_instance.cpp
    # NGCHW, GKYXC, NGKHW
+   xdl/mem/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_mem_inter_instance.cpp
    xdl/mem/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f16_mem_inter_instance.cpp
    xdl/mem/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f32_mem_inter_instance.cpp
    xdl/mem/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_int8_mem_inter_instance.cpp
@@ -56,6 +60,7 @@ add_instance_library(device_grouped_conv2d_fwd_instance
    xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_comp_instance.cpp
    xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_comp_instance.cpp
    # NGCHW, GKYXC, NGKHW
+   xdl/comp/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_comp_instance.cpp
    xdl/comp/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f16_comp_instance.cpp
    xdl/comp/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_f32_comp_instance.cpp
    xdl/comp/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_int8_comp_instance.cpp
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_comp_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_comp_instance.cpp
new file mode 100644
index 0000000000..65e233ce08
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_comp_instance.cpp
@@ -0,0 +1,39 @@
+// SPDX-License-Identifier: MIT
+// Copyright (c) 2024, Advanced Micro Devices, Inc. All rights reserved.
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_comp_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_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_comp_instances(
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
+                                                                NGCHW,
+                                                                GKYXC,
+                                                                Empty_Tuple,
+                                                                NGKHW,
+                                                                BF16,
+                                                                BF16,
+                                                                Empty_Tuple,
+                                                                BF16,
+                                                                PassThrough,
+                                                                PassThrough,
+                                                                PassThrough>>>& instances)
+{
+    add_device_operation_instances(
+        instances,
+        device_grouped_conv_fwd_xdl_bf16_comp_instances<2,
+                                                        NGCHW,
+                                                        GKYXC,
+                                                        Empty_Tuple,
+                                                        NGKHW,
+                                                        ConvFwdDefault>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_instance.cpp
new file mode 100644
index 0000000000..6ee6aa1e4a
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_instance.cpp
@@ -0,0 +1,38 @@
+// SPDX-License-Identifier: MIT
+// Copyright (c) 2024, Advanced Micro Devices, Inc. All rights reserved.
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_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_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_instances(
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
+                                                                NGCHW,
+                                                                GKYXC,
+                                                                Empty_Tuple,
+                                                                NGKHW,
+                                                                BF16,
+                                                                BF16,
+                                                                Empty_Tuple,
+                                                                BF16,
+                                                                PassThrough,
+                                                                PassThrough,
+                                                                PassThrough>>>& instances)
+{
+    add_device_operation_instances(instances,
+                                   device_grouped_conv_fwd_xdl_bf16_instances<2,
+                                                                              NGCHW,
+                                                                              GKYXC,
+                                                                              Empty_Tuple,
+                                                                              NGKHW,
+                                                                              ConvFwdDefault>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_mem_inter_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_mem_inter_instance.cpp
new file mode 100644
index 0000000000..88b5f30da5
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_mem_inter_instance.cpp
@@ -0,0 +1,39 @@
+// SPDX-License-Identifier: MIT
+// Copyright (c) 2024, Advanced Micro Devices, Inc. All rights reserved.
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_mem_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_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_mem_inter_instances(
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
+                                                                NGCHW,
+                                                                GKYXC,
+                                                                Empty_Tuple,
+                                                                NGKHW,
+                                                                BF16,
+                                                                BF16,
+                                                                Empty_Tuple,
+                                                                BF16,
+                                                                PassThrough,
+                                                                PassThrough,
+                                                                PassThrough>>>& instances)
+{
+    add_device_operation_instances(instances,
+                                   device_grouped_conv_fwd_xdl_bf16_mem_instances<2,
+                                                                                  NGCHW,
+                                                                                  GKYXC,
+                                                                                  Empty_Tuple,
+                                                                                  NGKHW,
+                                                                                  ConvFwdDefault,
+                                                                                  Interwave>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_mem_intra_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_mem_intra_instance.cpp
new file mode 100644
index 0000000000..48cca9c3f3
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_mem_intra_instance.cpp
@@ -0,0 +1,39 @@
+// SPDX-License-Identifier: MIT
+// Copyright (c) 2024, Advanced Micro Devices, Inc. All rights reserved.
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_mem_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_conv2d_fwd_xdl_ngchw_gkyxc_ngkhw_bf16_mem_intra_instances(
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
+                                                                NGCHW,
+                                                                GKYXC,
+                                                                Empty_Tuple,
+                                                                NGKHW,
+                                                                BF16,
+                                                                BF16,
+                                                                Empty_Tuple,
+                                                                BF16,
+                                                                PassThrough,
+                                                                PassThrough,
+                                                                PassThrough>>>& instances)
+{
+    add_device_operation_instances(instances,
+                                   device_grouped_conv_fwd_xdl_bf16_mem_instances<2,
+                                                                                  NGCHW,
+                                                                                  GKYXC,
+                                                                                  Empty_Tuple,
+                                                                                  NGKHW,
+                                                                                  ConvFwdDefault,
+                                                                                  Intrawave>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/merged_groups/device_grouped_conv2d_fwd_xdl_merged_groups_ngchw_gkyxc_ngkhw_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/merged_groups/device_grouped_conv2d_fwd_xdl_merged_groups_ngchw_gkyxc_ngkhw_bf16_instance.cpp
new file mode 100644
index 0000000000..14f00d8e88
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/merged_groups/device_grouped_conv2d_fwd_xdl_merged_groups_ngchw_gkyxc_ngkhw_bf16_instance.cpp
@@ -0,0 +1,48 @@
+// SPDX-License-Identifier: MIT
+// Copyright (c) 2024, Advanced Micro Devices, Inc. All rights reserved.
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_merged_groups_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_conv2d_fwd_xdl_merged_groups_ngchw_gkyxc_ngkhw_bf16_instances(
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
+                                                                NGCHW,
+                                                                GKYXC,
+                                                                Empty_Tuple,
+                                                                NGKHW,
+                                                                BF16,
+                                                                BF16,
+                                                                Empty_Tuple,
+                                                                BF16,
+                                                                PassThrough,
+                                                                PassThrough,
+                                                                PassThrough>>>& instances)
+{
+    add_device_operation_instances(
+        instances,
+        device_grouped_conv_fwd_xdl_merged_groups_bf16_instances<2,
+                                                                 NGCHW,
+                                                                 GKYXC,
+                                                                 Empty_Tuple,
+                                                                 NGKHW,
+                                                                 ConvFwdDefault>{});
+
+    add_device_operation_instances(
+        instances,
+        device_grouped_conv_fwd_xdl_merged_groups_bf16_instances<2,
+                                                                 NGCHW,
+                                                                 GKYXC,
+                                                                 Empty_Tuple,
+                                                                 NGKHW,
+                                                                 ConvFwd3x3>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck
diff --git a/test/grouped_convnd_fwd/test_grouped_convnd_fwd.cpp b/test/grouped_convnd_fwd/test_grouped_convnd_fwd.cpp
index 1abd4fd9f6..25481e0d7f 100644
--- a/test/grouped_convnd_fwd/test_grouped_convnd_fwd.cpp
+++ b/test/grouped_convnd_fwd/test_grouped_convnd_fwd.cpp
@@ -64,6 +64,7 @@ using KernelTypes2d = ::testing::Types<std::tuple<float, GNHWC, GKYXC, GNHWK>,
                                        std::tuple<int8_t, NHWGC, GKYXC, NHWGK>,
                                        std::tuple<float, NGCHW, GKYXC, NGKHW>,
                                        std::tuple<ck::half_t, NGCHW, GKYXC, NGKHW>,
+                                       std::tuple<ck::bhalf_t, NGCHW, GKYXC, NGKHW>,
                                        std::tuple<int8_t, NGCHW, GKYXC, NGKHW>>;
 
 using KernelTypes3d = ::testing::Types<std::tuple<float, GNDHWC, GKZYXC, GNDHWK>,