diff --git a/backends/vulkan/partitioner/supported_ops.py b/backends/vulkan/partitioner/supported_ops.py
index 08d7f96a6b..ca7ce72cae 100644
--- a/backends/vulkan/partitioner/supported_ops.py
+++ b/backends/vulkan/partitioner/supported_ops.py
@@ -8,7 +8,10 @@
 
 import operator
 
-from executorch.backends.vulkan.passes.custom_ops_defs import grid_priors_op  # noqa
+from executorch.backends.vulkan.passes.custom_ops_defs import (  # noqa
+    conv_with_clamp_op,
+    grid_priors_op,
+)
 
 from executorch.exir.dialects._ops import ops as exir_ops
 
@@ -84,6 +87,7 @@ def __contains__(self, op):
 
 CONVOLUTION_OPS = [
     exir_ops.edge.aten.convolution.default,
+    exir_ops.edge.et_vk.conv_with_clamp.default,
 ]
 
 REDUCTION_OPS = [
diff --git a/backends/vulkan/passes/custom_ops_defs.py b/backends/vulkan/passes/custom_ops_defs.py
index 62f21bfee6..fd586b665a 100644
--- a/backends/vulkan/passes/custom_ops_defs.py
+++ b/backends/vulkan/passes/custom_ops_defs.py
@@ -48,6 +48,43 @@ def conv_with_clamp_impl(
 conv_with_clamp_op = getattr(getattr(torch.ops, namespace), name)
 
 
+def conv_with_clamp_out_impl(
+    input,
+    weight,
+    bias=None,
+    stride=1,
+    padding=0,
+    dilation=1,
+    transposed=False,
+    output_padding=0,
+    groups=1,
+    output_min=-float("inf"),
+    output_max=float("inf"),
+    out=None,
+):
+    out = conv_with_clamp_impl(
+        input,
+        weight,
+        bias,
+        stride,
+        padding,
+        dilation,
+        transposed,
+        output_padding,
+        groups,
+        output_min,
+        output_max,
+    )
+    return out
+
+
+name = "conv_with_clamp.out"
+lib.define(
+    f"{name}(Tensor input, Tensor weight, Tensor? bias, SymInt[] stride, SymInt[] padding, SymInt[] dilation, bool transposed, SymInt[] output_padding, SymInt groups, Scalar? output_min, Scalar? output_max, *, Tensor(a!) out) -> Tensor(a!)"
+)
+lib.impl(name, conv_with_clamp_out_impl, "CompositeExplicitAutograd")
+
+
 # The dimension of x should be larger than 1
 def grid_priors_impl(
     x,
diff --git a/backends/vulkan/runtime/graph/ops/impl/Convolution.cpp b/backends/vulkan/runtime/graph/ops/impl/Convolution.cpp
index 52af0542b6..74113197d4 100644
--- a/backends/vulkan/runtime/graph/ops/impl/Convolution.cpp
+++ b/backends/vulkan/runtime/graph/ops/impl/Convolution.cpp
@@ -562,6 +562,7 @@ void conv(ComputeGraph& graph, const std::vector<ValueRef>& args) {
 REGISTER_OPERATORS {
   VK_REGISTER_OP(aten.convolution.default, conv);
   VK_REGISTER_OP(conv_with_clamp.default, conv);
+  VK_REGISTER_OP(et_vk.conv_with_clamp.default, conv);
 }
 
 } // namespace vkcompute
diff --git a/backends/vulkan/test/test_vulkan_delegate.py b/backends/vulkan/test/test_vulkan_delegate.py
index 9f57ec49a8..d80809ec79 100644
--- a/backends/vulkan/test/test_vulkan_delegate.py
+++ b/backends/vulkan/test/test_vulkan_delegate.py
@@ -1633,6 +1633,42 @@ def forward(self, x):
             memory_layouts=[vk_graph_schema.VkMemoryLayout.TENSOR_CHANNELS_PACKED],
         )
 
+    def test_vulkan_backend_conv_with_clamp(self):
+        class ConvWithClampModule(torch.nn.Module):
+            def __init__(self):
+                super().__init__()
+                self.weight = torch.randn(6, 8, 3, 3)
+                self.bias = torch.randn(8)
+                self.stride = (1, 2)
+                self.padding = (2, 3)
+                self.dilation = (1, 1)
+                self.transposed = True
+                self.output_padding = (0, 1)
+                self.groups = 1
+                self.output_min = 0
+                self.output_max = 10
+
+            def forward(self, x):
+                return torch.ops.et_vk.conv_with_clamp(
+                    x,
+                    self.weight,
+                    self.bias,
+                    self.stride,
+                    self.padding,
+                    self.dilation,
+                    self.transposed,
+                    self.output_padding,
+                    self.groups,
+                    self.output_min,
+                    self.output_max,
+                )
+
+        self.lower_module_and_test_output(
+            ConvWithClampModule(),
+            (torch.randn(size=(1, 6, 40, 50), dtype=torch.float32),),
+            memory_layouts=[vk_graph_schema.VkMemoryLayout.TENSOR_CHANNELS_PACKED],
+        )
+
     def test_vulkan_backend_grid_priors(self):
         class GridPriorsModule(torch.nn.Module):
             def __init__(self):