diff --git a/src/armnn/InternalTypes.cpp b/src/armnn/InternalTypes.cpp
index 9b46436..96af602 100644
--- a/src/armnn/InternalTypes.cpp
+++ b/src/armnn/InternalTypes.cpp
@@ -14,6 +14,7 @@
 {
     switch (type)
     {
+        case LayerType::Abs: return "Abs";
         case LayerType::Activation: return "Activation";
         case LayerType::Addition: return "Addition";
         case LayerType::BatchNormalization: return "BatchNormalization";
diff --git a/src/backends/neon/NeonLayerSupport.cpp b/src/backends/neon/NeonLayerSupport.cpp
index 80695fb..67c1560 100644
--- a/src/backends/neon/NeonLayerSupport.cpp
+++ b/src/backends/neon/NeonLayerSupport.cpp
@@ -18,6 +18,7 @@
 
 #if defined(ARMCOMPUTENEON_ENABLED)
 #include <aclCommon/ArmComputeUtils.hpp>
+#include "workloads/NeonAbsWorkload.hpp"
 #include "workloads/NeonAdditionWorkload.hpp"
 #include "workloads/NeonActivationWorkload.hpp"
 #include "workloads/NeonBatchNormalizationWorkload.hpp"
@@ -108,6 +109,16 @@
 
 } // anonymous namespace
 
+bool NeonLayerSupport::IsAbsSupported(const TensorInfo& input,
+                                      const TensorInfo& output,
+                                      Optional<std::string&> reasonIfUnsupported) const
+{
+    FORWARD_WORKLOAD_VALIDATE_FUNC(NeonAbsWorkloadValidate,
+                                   reasonIfUnsupported,
+                                   input,
+                                   output);
+}
+
 bool NeonLayerSupport::IsActivationSupported(const TensorInfo& input,
                                              const TensorInfo& output,
                                              const ActivationDescriptor& descriptor,
diff --git a/src/backends/neon/NeonLayerSupport.hpp b/src/backends/neon/NeonLayerSupport.hpp
index 7ba90e4..16a7030 100644
--- a/src/backends/neon/NeonLayerSupport.hpp
+++ b/src/backends/neon/NeonLayerSupport.hpp
@@ -12,6 +12,10 @@
 class NeonLayerSupport : public LayerSupportBase
 {
 public:
+    bool IsAbsSupported(const TensorInfo& input,
+                        const TensorInfo& output,
+                        Optional<std::string&> reasonIfUnsupported = EmptyOptional()) const override;
+
     bool IsActivationSupported(const TensorInfo& input,
                                const TensorInfo& output,
                                const ActivationDescriptor& descriptor,
diff --git a/src/backends/neon/NeonWorkloadFactory.cpp b/src/backends/neon/NeonWorkloadFactory.cpp
index a22a64f..fc1a65d 100644
--- a/src/backends/neon/NeonWorkloadFactory.cpp
+++ b/src/backends/neon/NeonWorkloadFactory.cpp
@@ -101,6 +101,12 @@
     return std::make_unique<CopyMemGenericWorkload>(descriptor, info);
 }
 
+std::unique_ptr<IWorkload> NeonWorkloadFactory::CreateAbs(const AbsQueueDescriptor& descriptor,
+                                                          const WorkloadInfo& info) const
+{
+    return std::make_unique<NeonAbsWorkload>(descriptor, info);
+}
+
 std::unique_ptr<IWorkload> NeonWorkloadFactory::CreateActivation(const ActivationQueueDescriptor& descriptor,
                                                                  const WorkloadInfo&              info) const
 {
diff --git a/src/backends/neon/NeonWorkloadFactory.hpp b/src/backends/neon/NeonWorkloadFactory.hpp
index de0bd5c..935f9e4 100644
--- a/src/backends/neon/NeonWorkloadFactory.hpp
+++ b/src/backends/neon/NeonWorkloadFactory.hpp
@@ -43,6 +43,9 @@
     std::unique_ptr<IWorkload> CreateOutput(const OutputQueueDescriptor& descriptor,
                                             const WorkloadInfo& info) const override;
 
+    std::unique_ptr<IWorkload> CreateAbs(const AbsQueueDescriptor& descriptor,
+                                         const WorkloadInfo& info) const override;
+
     std::unique_ptr<IWorkload> CreateActivation(const ActivationQueueDescriptor& descriptor,
                                                 const WorkloadInfo& info) const override;
 
diff --git a/src/backends/neon/backend.mk b/src/backends/neon/backend.mk
index e29919c..706e8b0 100644
--- a/src/backends/neon/backend.mk
+++ b/src/backends/neon/backend.mk
@@ -21,6 +21,7 @@
         NeonTensorHandleFactory.cpp \
         NeonTimer.cpp \
         NeonWorkloadFactory.cpp \
+        workloads/NeonAbsWorkload.cpp \
         workloads/NeonActivationWorkload.cpp \
         workloads/NeonAdditionWorkload.cpp \
         workloads/NeonBatchNormalizationWorkload.cpp \
diff --git a/src/backends/neon/test/NeonEndToEndTests.cpp b/src/backends/neon/test/NeonEndToEndTests.cpp
index 33c377f..eb41e94 100644
--- a/src/backends/neon/test/NeonEndToEndTests.cpp
+++ b/src/backends/neon/test/NeonEndToEndTests.cpp
@@ -5,6 +5,7 @@
 
 #include <backendsCommon/test/EndToEndTestImpl.hpp>
 
+#include <backendsCommon/test/AbsEndToEndTestImpl.hpp>
 #include <backendsCommon/test/ArithmeticTestImpl.hpp>
 #include <backendsCommon/test/ConcatEndToEndTestImpl.hpp>
 #include <backendsCommon/test/DequantizeEndToEndTestImpl.hpp>
@@ -20,6 +21,13 @@
 
 std::vector<armnn::BackendId> defaultBackends = {armnn::Compute::CpuAcc};
 
+// Abs
+BOOST_AUTO_TEST_CASE(NeonAbsEndToEndTestFloat32)
+{
+    AbsEndToEnd<armnn::DataType::Float32>(defaultBackends);
+}
+
+// Constant
 BOOST_AUTO_TEST_CASE(ConstantUsage_Neon_Float32)
 {
     BOOST_TEST(ConstantUsageFloat32Test(defaultBackends));
diff --git a/src/backends/neon/test/NeonLayerTests.cpp b/src/backends/neon/test/NeonLayerTests.cpp
index e0f2e84..5a0c966 100644
--- a/src/backends/neon/test/NeonLayerTests.cpp
+++ b/src/backends/neon/test/NeonLayerTests.cpp
@@ -853,6 +853,12 @@
                      MultiChannelTransposeConvolution2dTest<DataType::QuantisedAsymm8, DataType::Signed32>,
                      DataLayout::NHWC)
 
+// Abs
+ARMNN_AUTO_TEST_CASE(Abs2d, Abs2dTest<DataType::Float32>)
+ARMNN_AUTO_TEST_CASE(Abs3d, Abs3dTest<DataType::Float32>)
+
+ARMNN_AUTO_TEST_CASE(AbsZero, AbsZeroTest<DataType::Float32>)
+
 #if defined(ARMNNREF_ENABLED)
 
 // The ARMNN_COMPARE_REF_AUTO_TEST_CASE and the ARMNN_COMPARE_REF_FIXTURE_TEST_CASE test units are not available
diff --git a/src/backends/neon/workloads/CMakeLists.txt b/src/backends/neon/workloads/CMakeLists.txt
index e03a713..e0831be 100644
--- a/src/backends/neon/workloads/CMakeLists.txt
+++ b/src/backends/neon/workloads/CMakeLists.txt
@@ -4,6 +4,8 @@
 #
 
 list(APPEND armnnNeonBackendWorkloads_sources
+    NeonAbsWorkload.cpp
+    NeonAbsWorkload.hpp
     NeonActivationWorkload.cpp
     NeonActivationWorkload.hpp
     NeonAdditionWorkload.cpp
diff --git a/src/backends/neon/workloads/NeonAbsWorkload.cpp b/src/backends/neon/workloads/NeonAbsWorkload.cpp
new file mode 100644
index 0000000..7f8ed5a
--- /dev/null
+++ b/src/backends/neon/workloads/NeonAbsWorkload.cpp
@@ -0,0 +1,43 @@
+//
+// Copyright © 2017 Arm Ltd. All rights reserved.
+// SPDX-License-Identifier: MIT
+//
+
+#include "NeonAbsWorkload.hpp"
+
+#include "NeonWorkloadUtils.hpp"
+
+#include <aclCommon/ArmComputeTensorHandle.hpp>
+#include <aclCommon/ArmComputeTensorUtils.hpp>
+
+#include <boost/cast.hpp>
+
+namespace armnn
+{
+
+arm_compute::Status NeonAbsWorkloadValidate(const TensorInfo& input, const TensorInfo& output)
+{
+    const arm_compute::TensorInfo aclInput  = armcomputetensorutils::BuildArmComputeTensorInfo(input);
+    const arm_compute::TensorInfo aclOutput = armcomputetensorutils::BuildArmComputeTensorInfo(output);
+
+    return arm_compute::NEAbsLayer::validate(&aclInput, &aclOutput);
+}
+
+NeonAbsWorkload::NeonAbsWorkload(const AbsQueueDescriptor& descriptor, const WorkloadInfo& info)
+    : BaseWorkload<AbsQueueDescriptor>(descriptor, info)
+{
+    m_Data.ValidateInputsOutputs("NeonAbsWorkload", 1, 1);
+
+    arm_compute::ITensor& input  = boost::polymorphic_downcast<IAclTensorHandle*>(m_Data.m_Inputs[0])->GetTensor();
+    arm_compute::ITensor& output = boost::polymorphic_downcast<IAclTensorHandle*>(m_Data.m_Outputs[0])->GetTensor();
+
+    m_AbsLayer.configure(&input, &output);
+}
+
+void NeonAbsWorkload::Execute() const
+{
+    ARMNN_SCOPED_PROFILING_EVENT_NEON("NeonAbsWorkload_Execute");
+    m_AbsLayer.run();
+}
+
+} // namespace armnn
diff --git a/src/backends/neon/workloads/NeonAbsWorkload.hpp b/src/backends/neon/workloads/NeonAbsWorkload.hpp
new file mode 100644
index 0000000..0024d19
--- /dev/null
+++ b/src/backends/neon/workloads/NeonAbsWorkload.hpp
@@ -0,0 +1,28 @@
+//
+// Copyright © 2017 Arm Ltd. All rights reserved.
+// SPDX-License-Identifier: MIT
+//
+
+#pragma once
+
+#include <backendsCommon/Workload.hpp>
+
+#include <arm_compute/core/Error.h>
+#include <arm_compute/runtime/NEON/functions/NEElementwiseUnaryLayer.h>
+
+namespace armnn
+{
+
+arm_compute::Status NeonAbsWorkloadValidate(const TensorInfo& input, const TensorInfo& output);
+
+class NeonAbsWorkload : public BaseWorkload<AbsQueueDescriptor>
+{
+public:
+    NeonAbsWorkload(const AbsQueueDescriptor& descriptor, const WorkloadInfo& info);
+    virtual void Execute() const override;
+
+private:
+    mutable arm_compute::NEAbsLayer m_AbsLayer;
+};
+
+} // namespace armnn
diff --git a/src/backends/neon/workloads/NeonWorkloads.hpp b/src/backends/neon/workloads/NeonWorkloads.hpp
index 210ba96..d807e80 100644
--- a/src/backends/neon/workloads/NeonWorkloads.hpp
+++ b/src/backends/neon/workloads/NeonWorkloads.hpp
@@ -4,6 +4,7 @@
 //
 
 #pragma once
+#include "NeonAbsWorkload.hpp"
 #include "NeonActivationWorkload.hpp"
 #include "NeonAdditionWorkload.hpp"
 #include "NeonBatchNormalizationWorkload.hpp"