diff --git a/src/backends/neon/NeonLayerSupport.cpp b/src/backends/neon/NeonLayerSupport.cpp
index cd4dca8..d097240 100644
--- a/src/backends/neon/NeonLayerSupport.cpp
+++ b/src/backends/neon/NeonLayerSupport.cpp
@@ -39,6 +39,7 @@
 #include "workloads/NeonDepthToSpaceWorkload.hpp"
 #include "workloads/NeonDepthwiseConvolutionWorkload.hpp"
 #include "workloads/NeonDequantizeWorkload.hpp"
+#include "workloads/NeonElementwiseBinaryWorkload.hpp"
 #include "workloads/NeonExpWorkload.hpp"
 #include "workloads/NeonInstanceNormalizationWorkload.hpp"
 #include "workloads/NeonL2NormalizationFloatWorkload.hpp"
@@ -387,6 +388,15 @@
                                                    infos[1],
                                                    infos[2],
                                                    nullptr);
+                case BinaryOperation::Power:
+                case BinaryOperation::SqDiff:
+                    FORWARD_WORKLOAD_VALIDATE_FUNC(NeonElementwiseBinaryWorkloadValidate,
+                                                   reasonIfUnsupported,
+                                                   infos[0],
+                                                   infos[1],
+                                                   infos[2],
+                                                   desc,
+                                                   nullptr);
                 case BinaryOperation::Sub:
                     FORWARD_WORKLOAD_VALIDATE_FUNC(NeonSubtractionWorkloadValidate,
                                                    reasonIfUnsupported,
diff --git a/src/backends/neon/NeonWorkloadFactory.cpp b/src/backends/neon/NeonWorkloadFactory.cpp
index c78b58d..eca3867 100644
--- a/src/backends/neon/NeonWorkloadFactory.cpp
+++ b/src/backends/neon/NeonWorkloadFactory.cpp
@@ -334,6 +334,11 @@
                     multiplicationQueueDescriptor.m_Outputs = descriptor.m_Outputs;
                     return std::make_unique<NeonMultiplicationWorkload>(multiplicationQueueDescriptor, info);
                 }
+                case BinaryOperation::Power:
+                case BinaryOperation::SqDiff:
+                {
+                    return std::make_unique<NeonElementwiseBinaryWorkload>(*elementwiseBinaryQueueDescriptor, info);
+                }
                 case BinaryOperation::Sub:
                 {
                     SubtractionQueueDescriptor subtractionQueueDescriptor;
diff --git a/src/backends/neon/backend.mk b/src/backends/neon/backend.mk
index bbc5554..e2439ee 100644
--- a/src/backends/neon/backend.mk
+++ b/src/backends/neon/backend.mk
@@ -1,5 +1,5 @@
 #
-# Copyright © 2017 ARM Ltd and Contributors. All rights reserved.
+# Copyright © 2017,2023 ARM Ltd and Contributors. All rights reserved.
 # SPDX-License-Identifier: MIT
 #
 
@@ -42,6 +42,7 @@
         workloads/NeonDepthwiseConvolutionWorkload.cpp \
         workloads/NeonDequantizeWorkload.cpp \
         workloads/NeonDetectionPostProcessWorkload.cpp \
+        workloads/NeonElementwiseBinaryWorkload.cpp \
         workloads/NeonExpWorkload.cpp \
         workloads/NeonFillWorkload.cpp \
         workloads/NeonFloorFloatWorkload.cpp \
diff --git a/src/backends/neon/test/NeonEndToEndTests.cpp b/src/backends/neon/test/NeonEndToEndTests.cpp
index fb05cc4..071ee41 100644
--- a/src/backends/neon/test/NeonEndToEndTests.cpp
+++ b/src/backends/neon/test/NeonEndToEndTests.cpp
@@ -14,6 +14,7 @@
 #include <backendsCommon/test/DepthToSpaceEndToEndTestImpl.hpp>
 #include <backendsCommon/test/DequantizeEndToEndTestImpl.hpp>
 #include <backendsCommon/test/DetectionPostProcessEndToEndTestImpl.hpp>
+#include <backendsCommon/test/ElementwiseBinaryEndToEndTestImpl.hpp>
 #include <backendsCommon/test/ElementwiseUnaryEndToEndTestImpl.hpp>
 #include <backendsCommon/test/FillEndToEndTestImpl.hpp>
 #include <backendsCommon/test/InstanceNormalizationEndToEndTestImpl.hpp>
@@ -133,6 +134,8 @@
                                                                   expectedOutput);
 }
 
+// ElementwiseBinary
+// Add
 TEST_CASE("NeonAdditionEndToEndFloat32Test")
 {
     AdditionEndToEnd<armnn::DataType::Float32>(neonDefaultBackends);
@@ -143,6 +146,24 @@
     AdditionEndToEnd<armnn::DataType::QAsymmU8>(neonDefaultBackends);
 }
 
+// Power
+TEST_CASE("RefPowerEndToEndTestFloat32")
+{
+    ElementwiseBinarySimpleEndToEnd<armnn::DataType::Float32>(neonDefaultBackends, BinaryOperation::Power);
+}
+
+// SqDiff
+TEST_CASE("RefSquaredDifferenceEndToEndTestFloat32")
+{
+    ElementwiseBinarySimpleEndToEnd<armnn::DataType::Float32>(neonDefaultBackends, BinaryOperation::SqDiff);
+}
+
+TEST_CASE("RefSquaredDifferenceEndToEndTestUint8")
+{
+    ElementwiseBinarySimpleEndToEnd<armnn::DataType::QAsymmU8>(neonDefaultBackends, BinaryOperation::SqDiff);
+}
+
+// Batch Mat Mul
 TEST_CASE("NeonBatchMatMulEndToEndFloat32Test")
 {
     BatchMatMulEndToEnd<armnn::DataType::Float32>(neonDefaultBackends);
@@ -153,6 +174,7 @@
     BatchMatMulEndToEnd<armnn::DataType::QAsymmS8>(neonDefaultBackends);
 }
 
+// Concat
 TEST_CASE("NeonConcatEndToEndDim0Test")
 {
     ConcatDim0EndToEnd<armnn::DataType::Float32>(neonDefaultBackends);
diff --git a/src/backends/neon/test/NeonLayerTests.cpp b/src/backends/neon/test/NeonLayerTests.cpp
index 7150607..c0b995b 100644
--- a/src/backends/neon/test/NeonLayerTests.cpp
+++ b/src/backends/neon/test/NeonLayerTests.cpp
@@ -817,6 +817,24 @@
 ARMNN_AUTO_TEST_CASE_WITH_THF(MultiplicationBroadcast1DVectorUint8, MultiplicationBroadcast1DVectorUint8Test)
 ARMNN_AUTO_TEST_CASE_WITH_THF(Multiplication5d, Multiplication5dTest)
 
+// SquaredDifference
+ARMNN_AUTO_TEST_CASE_WITH_THF(SimpleSquaredDifference, SquaredDifferenceTest)
+ARMNN_AUTO_TEST_CASE_WITH_THF(SquaredDiffBroadcast1Element, SquaredDiffBroadcast1ElementTest)
+ARMNN_AUTO_TEST_CASE_WITH_THF(SquaredDiffBroadcast, SquaredDiffBroadcastTest)
+
+ARMNN_AUTO_TEST_CASE_WITH_THF(SquaredDifferenceUint8, SquaredDifferenceUint8Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(SquaredDiffBroadcastUint8, SquaredDiffBroadcastUint8Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(SquaredDiffBroadcast1ElementUint8, SquaredDiffBroadcast1ElementUint8Test)
+
+ARMNN_AUTO_TEST_CASE_WITH_THF(SquaredDifferenceInt32, SquaredDifferenceInt32Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(SquaredDiffBroadcastInt32, SquaredDiffBroadcastInt32Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(SquaredDiffBroadcast1ElementInt32, SquaredDiffBroadcast1ElementInt32Test)
+
+// Power
+ARMNN_AUTO_TEST_CASE_WITH_THF(SimplePower, PowerTest)
+ARMNN_AUTO_TEST_CASE_WITH_THF(PowerBroadcast1Element, PowerBroadcast1ElementTest)
+ARMNN_AUTO_TEST_CASE_WITH_THF(PowerBroadcast, PowerBroadcastTest)
+
 // Batch Norm
 ARMNN_AUTO_TEST_CASE_WITH_THF(BatchNormFloat32, BatchNormFloat32Test)
 ARMNN_AUTO_TEST_CASE_WITH_THF(BatchNormFloat32Nhwc, BatchNormFloat32NhwcTest)
diff --git a/src/backends/neon/workloads/CMakeLists.txt b/src/backends/neon/workloads/CMakeLists.txt
index a3eb883..f3f1284 100644
--- a/src/backends/neon/workloads/CMakeLists.txt
+++ b/src/backends/neon/workloads/CMakeLists.txt
@@ -1,5 +1,5 @@
 #
-# Copyright © 2017 Arm Ltd and Contributors. All rights reserved.
+# Copyright © 2017,2023 Arm Ltd and Contributors. All rights reserved.
 # SPDX-License-Identifier: MIT
 #
 
@@ -44,6 +44,8 @@
     NeonDequantizeWorkload.hpp
     NeonDetectionPostProcessWorkload.cpp
     NeonDetectionPostProcessWorkload.hpp
+    NeonElementwiseBinaryWorkload.cpp
+    NeonElementwiseBinaryWorkload.hpp
     NeonExpWorkload.cpp
     NeonExpWorkload.hpp
     NeonFillWorkload.cpp
diff --git a/src/backends/neon/workloads/NeonElementwiseBinaryWorkload.cpp b/src/backends/neon/workloads/NeonElementwiseBinaryWorkload.cpp
new file mode 100644
index 0000000..839ec6c
--- /dev/null
+++ b/src/backends/neon/workloads/NeonElementwiseBinaryWorkload.cpp
@@ -0,0 +1,92 @@
+//
+// Copyright © 2023 Arm Ltd and Contributors. All rights reserved.
+// SPDX-License-Identifier: MIT
+//
+
+#include "NeonElementwiseBinaryWorkload.hpp"
+#include "NeonWorkloadUtils.hpp"
+
+#include <aclCommon/ArmComputeTensorUtils.hpp>
+#include <aclCommon/ArmComputeUtils.hpp>
+
+#include <armnn/utility/PolymorphicDowncast.hpp>
+#include <armnn/backends/TensorHandle.hpp>
+
+#include <arm_compute/runtime/NEON/functions/NEElementwiseOperations.h>
+
+namespace armnn
+{
+
+arm_compute::Status NeonElementwiseBinaryWorkloadValidate(const TensorInfo& input0,
+                                                          const TensorInfo& input1,
+                                                          const TensorInfo& output,
+                                                          const ElementwiseBinaryDescriptor& descriptor,
+                                                          const ActivationDescriptor* activationDescriptor)
+{
+    const arm_compute::TensorInfo aclInput0 = armcomputetensorutils::BuildArmComputeTensorInfo(input0);
+    const arm_compute::TensorInfo aclInput1 = armcomputetensorutils::BuildArmComputeTensorInfo(input1);
+    const arm_compute::TensorInfo aclOutput = armcomputetensorutils::BuildArmComputeTensorInfo(output);
+
+    const arm_compute::ActivationLayerInfo activationInfo = ConvertActivationDescriptorToAclActivationLayerInfo(
+            activationDescriptor);
+
+    switch (descriptor.m_Operation)
+    {
+        case armnn::BinaryOperation::Power:
+            return arm_compute::NEElementwisePower::validate(&aclInput0,
+                                                             &aclInput1,
+                                                             &aclOutput,
+                                                             activationInfo);
+        case armnn::BinaryOperation::SqDiff:
+            return arm_compute::NEElementwiseSquaredDiff::validate(&aclInput0,
+                                                                   &aclInput1,
+                                                                   &aclOutput,
+                                                                   activationInfo);
+        default:
+            throw InvalidArgumentException("Unknown binary operator", CHECK_LOCATION());
+    }
+}
+
+
+NeonElementwiseBinaryWorkload::NeonElementwiseBinaryWorkload(const ElementwiseBinaryQueueDescriptor& descriptor,
+                                                             const WorkloadInfo& info)
+    : NeonBaseWorkload<ElementwiseBinaryQueueDescriptor>(descriptor, info)
+{
+    m_Data.ValidateInputsOutputs("NeonElementwiseBinaryWorkload", 2, 1);
+
+    arm_compute::ITensor& input1 = PolymorphicDowncast<IAclTensorHandle*>(m_Data.m_Inputs[0])->GetTensor();
+    arm_compute::ITensor& input2 = PolymorphicDowncast<IAclTensorHandle*>(m_Data.m_Inputs[1])->GetTensor();
+    arm_compute::ITensor& output = PolymorphicDowncast<IAclTensorHandle*>(m_Data.m_Outputs[0])->GetTensor();
+
+    const arm_compute::ActivationLayerInfo activationInfo = ConvertAdditionalInfoToAclActivationLayerInfo(descriptor);
+
+    ARMNN_SCOPED_PROFILING_EVENT(Compute::Undefined, "NeonElementwiseBinaryWorkload_configure");
+
+    switch (descriptor.m_Parameters.m_Operation)
+    {
+        case armnn::BinaryOperation::Power:
+        {
+            auto powerLayer = std::make_unique<arm_compute::NEElementwisePower>();
+            powerLayer->configure(&input1, &input2, &output, activationInfo);
+            m_ElementwiseBinaryLayer.reset(powerLayer.release());
+            break;
+        }
+        case armnn::BinaryOperation::SqDiff:
+        {
+            auto SqDiffLayer = std::make_unique<arm_compute::NEElementwiseSquaredDiff>();
+            SqDiffLayer->configure(&input1, &input2, &output, activationInfo);
+            m_ElementwiseBinaryLayer.reset(SqDiffLayer.release());
+            break;
+        }
+        default:
+            throw InvalidArgumentException("Unknown binary operator", CHECK_LOCATION());
+    }
+}
+
+void NeonElementwiseBinaryWorkload::Execute() const
+{
+    ARMNN_SCOPED_PROFILING_EVENT_NEON_GUID("NeonElementwiseBinaryWorkload_Execute", this->GetGuid());
+    m_ElementwiseBinaryLayer->run();
+}
+
+} //namespace armnn
\ No newline at end of file
diff --git a/src/backends/neon/workloads/NeonElementwiseBinaryWorkload.hpp b/src/backends/neon/workloads/NeonElementwiseBinaryWorkload.hpp
new file mode 100644
index 0000000..9cdc7ae
--- /dev/null
+++ b/src/backends/neon/workloads/NeonElementwiseBinaryWorkload.hpp
@@ -0,0 +1,33 @@
+//
+// Copyright © 2023 Arm Ltd and Contributors. All rights reserved.
+// SPDX-License-Identifier: MIT
+//
+
+#pragma once
+
+#include "NeonBaseWorkload.hpp"
+
+#include <arm_compute/core/Error.h>
+#include <arm_compute/core/Types.h>
+#include <arm_compute/runtime/IFunction.h>
+
+namespace armnn
+{
+
+arm_compute::Status NeonElementwiseBinaryWorkloadValidate(const TensorInfo& input0,
+                                                          const TensorInfo& input1,
+                                                          const TensorInfo& output,
+                                                          const ElementwiseBinaryDescriptor& descriptor,
+                                                          const ActivationDescriptor* activationDescriptor = nullptr);
+
+class NeonElementwiseBinaryWorkload : public NeonBaseWorkload<ElementwiseBinaryQueueDescriptor>
+{
+public:
+    NeonElementwiseBinaryWorkload(const ElementwiseBinaryQueueDescriptor& descriptor, const WorkloadInfo& info);
+    virtual void Execute() const override;
+
+private:
+    std::unique_ptr<arm_compute::IFunction> m_ElementwiseBinaryLayer;
+};
+
+} //namespace armnn
\ No newline at end of file
diff --git a/src/backends/neon/workloads/NeonWorkloads.hpp b/src/backends/neon/workloads/NeonWorkloads.hpp
index 01fd2f7..0247486 100644
--- a/src/backends/neon/workloads/NeonWorkloads.hpp
+++ b/src/backends/neon/workloads/NeonWorkloads.hpp
@@ -1,5 +1,5 @@
 //
-// Copyright © 2017 Arm Ltd and Contributors. All rights reserved.
+// Copyright © 2017, 2023 Arm Ltd and Contributors. All rights reserved.
 // SPDX-License-Identifier: MIT
 //
 
@@ -25,6 +25,7 @@
 #include "NeonDequantizeWorkload.hpp"
 #include "NeonDetectionPostProcessWorkload.hpp"
 #include "NeonDivisionWorkload.hpp"
+#include "NeonElementwiseBinaryWorkload.hpp"
 #include "NeonExpWorkload.hpp"
 #include "NeonFillWorkload.hpp"
 #include "NeonFloorFloatWorkload.hpp"