diff --git a/src/backends/reference/test/RefEndToEndTests.cpp b/src/backends/reference/test/RefEndToEndTests.cpp
index 95004c4..4bb3f29 100644
--- a/src/backends/reference/test/RefEndToEndTests.cpp
+++ b/src/backends/reference/test/RefEndToEndTests.cpp
@@ -1602,6 +1602,22 @@
 {
     ElementwiseBinarySimpleEndToEnd<armnn::DataType::QAsymmU8>(defaultBackends, BinaryOperation::Minimum);
 }
+TEST_CASE("RefPowerEndToEndTestFloat32")
+{
+    ElementwiseBinarySimpleEndToEnd<armnn::DataType::Float32>(defaultBackends, BinaryOperation::Power);
+}
+TEST_CASE("RefPowerEndToEndTestUint8")
+{
+    ElementwiseBinarySimpleEndToEnd<armnn::DataType::QAsymmU8>(defaultBackends, BinaryOperation::Power);
+}
+TEST_CASE("RefSquaredDifferenceEndToEndTestFloat32")
+{
+    ElementwiseBinarySimpleEndToEnd<armnn::DataType::Float32>(defaultBackends, BinaryOperation::SqDiff);
+}
+TEST_CASE("RefSquaredDifferenceEndToEndTestUint8")
+{
+    ElementwiseBinarySimpleEndToEnd<armnn::DataType::QAsymmU8>(defaultBackends, BinaryOperation::SqDiff);
+}
 #endif
 
 }
diff --git a/src/backends/reference/test/RefLayerTests.cpp b/src/backends/reference/test/RefLayerTests.cpp
index 0e228db..8b89743 100644
--- a/src/backends/reference/test/RefLayerTests.cpp
+++ b/src/backends/reference/test/RefLayerTests.cpp
@@ -1,5 +1,5 @@
 //
-// Copyright © 2017,2022 Arm Ltd and Contributors. All rights reserved.
+// Copyright © 2017,2022, 2023 Arm Ltd and Contributors. All rights reserved.
 // SPDX-License-Identifier: MIT
 //
 
@@ -1003,6 +1003,48 @@
 ARMNN_AUTO_TEST_CASE_WITH_THF(MultiplicationBroadcast1DVectorInt32, MultiplicationBroadcast1DVectorInt32Test)
 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(SquaredDifferenceFloat16, SquaredDifferenceFloat16Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(SquaredDiffBroadcast1ElementFloat16, SquaredDiffBroadcast1ElementFloat16Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(SquaredDiffBroadcastFloat16, SquaredDiffBroadcastFloat16Test)
+
+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(SquaredDifferenceInt16, SquaredDifferenceInt16Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(SquaredDiffBroadcastInt16, SquaredDiffBroadcastInt16Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(SquaredDiffBroadcast1ElementInt16, SquaredDiffBroadcast1ElementInt16Test)
+
+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)
+
+ARMNN_AUTO_TEST_CASE_WITH_THF(PowerFloat16, PowerFloat16Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(PowerBroadcast1ElementFloat16, PowerBroadcast1ElementFloat16Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(PowerBroadcastFloat16, PowerBroadcastFloat16Test)
+
+ARMNN_AUTO_TEST_CASE_WITH_THF(PowerUint8, PowerUint8Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(PowerBroadcastUint8, PowerBroadcastUint8Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(PowerBroadcast1ElementUint8, PowerBroadcast1ElementUint8Test)
+
+ARMNN_AUTO_TEST_CASE_WITH_THF(PowerInt16, PowerInt16Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(PowerBroadcastInt16, PowerBroadcastInt16Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(PowerBroadcast1ElementInt16, PowerBroadcast1ElementInt16Test)
+
+ARMNN_AUTO_TEST_CASE_WITH_THF(PowerInt32, PowerInt32Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(PowerBroadcastInt32, PowerBroadcastInt32Test)
+ARMNN_AUTO_TEST_CASE_WITH_THF(PowerBroadcast1ElementInt32, PowerBroadcast1ElementInt32Test)
+
 // Batch Mat Mul
 ARMNN_AUTO_TEST_CASE_WITH_THF(BatchMatMul2DSimpleFloat32, BatchMatMul2DSimpleTest<DataType::Float32>);
 ARMNN_AUTO_TEST_CASE_WITH_THF(BatchMatMul2DSimpleFloat16, BatchMatMul2DSimpleTest<DataType::Float16>);
diff --git a/src/backends/reference/workloads/ElementwiseFunction.cpp b/src/backends/reference/workloads/ElementwiseFunction.cpp
index c5b0ad1..4044f06 100644
--- a/src/backends/reference/workloads/ElementwiseFunction.cpp
+++ b/src/backends/reference/workloads/ElementwiseFunction.cpp
@@ -14,6 +14,8 @@
 #include "Rsqrt.hpp"
 #include "Sin.hpp"
 #include "Sqrt.hpp"
+#include "Power.hpp"
+#include "SquaredDifference.hpp"
 
 
 namespace armnn
@@ -67,6 +69,8 @@
 template struct armnn::ElementwiseBinaryFunction<std::divides<float>>;
 template struct armnn::ElementwiseBinaryFunction<armnn::maximum<float>>;
 template struct armnn::ElementwiseBinaryFunction<armnn::minimum<float>>;
+template struct armnn::ElementwiseBinaryFunction<armnn::power<float>>;
+template struct armnn::ElementwiseBinaryFunction<armnn::squaredDifference<float>>;
 
 template struct armnn::ElementwiseBinaryFunction<std::plus<int32_t>>;
 template struct armnn::ElementwiseBinaryFunction<std::minus<int32_t>>;
@@ -74,6 +78,8 @@
 template struct armnn::ElementwiseBinaryFunction<std::divides<int32_t>>;
 template struct armnn::ElementwiseBinaryFunction<armnn::maximum<int32_t>>;
 template struct armnn::ElementwiseBinaryFunction<armnn::minimum<int32_t>>;
+template struct armnn::ElementwiseBinaryFunction<armnn::power<int32_t>>;
+template struct armnn::ElementwiseBinaryFunction<armnn::squaredDifference<int32_t>>;
 
 // Comparison
 template struct armnn::ElementwiseBinaryFunction<std::equal_to<float>>;
diff --git a/src/backends/reference/workloads/Power.hpp b/src/backends/reference/workloads/Power.hpp
new file mode 100644
index 0000000..744328e
--- /dev/null
+++ b/src/backends/reference/workloads/Power.hpp
@@ -0,0 +1,27 @@
+//
+// Copyright © 2023 Arm Ltd and Contributors. All rights reserved.
+// SPDX-License-Identifier: MIT
+//
+
+#pragma once
+
+#include <iostream>
+
+namespace armnn
+{
+
+template<typename T>
+struct power
+{
+    typedef T result_type;
+    typedef T first_argument_type;
+
+    T
+    operator()(const T& input1, const T& input2) const
+    {
+        T power = armnn::numeric_cast<T>(std::pow(static_cast<float>(input1), static_cast<float>(input2)));
+        return power;
+    }
+};
+
+} //namespace armnn
diff --git a/src/backends/reference/workloads/RefElementwiseBinaryWorkload.cpp b/src/backends/reference/workloads/RefElementwiseBinaryWorkload.cpp
index 5dc77f8..e71cdd4 100644
--- a/src/backends/reference/workloads/RefElementwiseBinaryWorkload.cpp
+++ b/src/backends/reference/workloads/RefElementwiseBinaryWorkload.cpp
@@ -11,6 +11,8 @@
 #include "RefWorkloadUtils.hpp"
 #include "Maximum.hpp"
 #include "Minimum.hpp"
+#include "SquaredDifference.hpp"
+#include "Power.hpp"
 
 #include <Profiling.hpp>
 
@@ -44,6 +46,8 @@
     using MinimumFunction = ElementwiseBinaryFunction<armnn::minimum<DataType>>;
     using MulFunction     = ElementwiseBinaryFunction<std::multiplies<DataType>>;
     using SubFunction     = ElementwiseBinaryFunction<std::minus<DataType>>;
+    using SqDiffFunction  = ElementwiseBinaryFunction<armnn::squaredDifference<DataType>>;
+    using PowerFunction   = ElementwiseBinaryFunction<armnn::power<DataType>>;
 
     switch (operation)
     {
@@ -77,6 +81,16 @@
             SubFunction(inShape0, inShape1, outShape, *input0, *input1, *output);
             break;
         }
+        case BinaryOperation::SqDiff:
+        {
+            SqDiffFunction(inShape0, inShape1, outShape, *input0, *input1, *output);
+            break;
+        }
+        case BinaryOperation::Power:
+        {
+            PowerFunction(inShape0, inShape1, outShape, *input0, *input1, *output);
+            break;
+        }
         default:
         {
             throw InvalidArgumentException(std::string("Unsupported binary operation ") +
diff --git a/src/backends/reference/workloads/SquaredDifference.hpp b/src/backends/reference/workloads/SquaredDifference.hpp
new file mode 100644
index 0000000..c15b379
--- /dev/null
+++ b/src/backends/reference/workloads/SquaredDifference.hpp
@@ -0,0 +1,28 @@
+//
+// Copyright © 2023 Arm Ltd and Contributors. All rights reserved.
+// SPDX-License-Identifier: MIT
+//
+
+#pragma once
+
+#include <cmath>
+
+namespace armnn
+{
+
+template<typename T>
+struct squaredDifference
+{
+    typedef T result_type;
+    typedef T first_argument_type;
+
+    T
+    operator()(const T& input1, const T& input2) const
+    {
+        float diff = std::minus<>{}(static_cast<float>(input1),static_cast<float>(input2));
+        T squaredDiff = armnn::numeric_cast<T>(std::pow(static_cast<float>(diff), 2));
+        return squaredDiff;
+    }
+};
+
+} //namespace armnn