diff --git a/src/backends/cl/workloads/CMakeLists.txt b/src/backends/cl/workloads/CMakeLists.txt
index 40c5f76..d98956f 100644
--- a/src/backends/cl/workloads/CMakeLists.txt
+++ b/src/backends/cl/workloads/CMakeLists.txt
@@ -70,6 +70,8 @@
     ClSoftmaxUint8Workload.hpp
     ClSpaceToBatchNdWorkload.hpp
     ClSpaceToBatchNdWorkload.cpp
+    ClSpaceToDepthWorkload.cpp
+    ClSpaceToDepthWorkload.hpp
     ClSplitterWorkload.cpp
     ClSplitterWorkload.hpp
     ClStridedSliceWorkload.cpp
diff --git a/src/backends/cl/workloads/ClSpaceToDepthWorkload.cpp b/src/backends/cl/workloads/ClSpaceToDepthWorkload.cpp
new file mode 100644
index 0000000..d541e4e
--- /dev/null
+++ b/src/backends/cl/workloads/ClSpaceToDepthWorkload.cpp
@@ -0,0 +1,61 @@
+//
+// Copyright © 2017 Arm Ltd. All rights reserved.
+// SPDX-License-Identifier: MIT
+//
+
+#include "ClSpaceToDepthWorkload.hpp"
+#include "ClWorkloadUtils.hpp"
+
+#include <aclCommon/ArmComputeUtils.hpp>
+#include <aclCommon/ArmComputeTensorUtils.hpp>
+#include <backendsCommon/CpuTensorHandle.hpp>
+#include <cl/ClTensorHandle.hpp>
+#include <boost/polymorphic_pointer_cast.hpp>
+
+namespace armnn
+{
+using namespace armcomputetensorutils;
+
+ClSpaceToDepthWorkload::ClSpaceToDepthWorkload(const SpaceToDepthQueueDescriptor& desc,
+                                               const WorkloadInfo& info)
+    : BaseWorkload<SpaceToDepthQueueDescriptor>(desc, info)
+{
+    m_Data.ValidateInputsOutputs("ClSpaceToDepthWorkload", 1, 1);
+
+    arm_compute::DataLayout aclDataLayout = ConvertDataLayout(m_Data.m_Parameters.m_DataLayout);
+
+    arm_compute::ICLTensor& input = static_cast<IClTensorHandle*>(m_Data.m_Inputs[0])->GetTensor();
+    input.info()->set_data_layout(aclDataLayout);
+
+    int32_t blockSize = boost::numeric_cast<int32_t>(desc.m_Parameters.m_BlockSize);
+
+    arm_compute::ICLTensor& output = static_cast<IClTensorHandle*>(m_Data.m_Outputs[0])->GetTensor();
+    output.info()->set_data_layout(aclDataLayout);
+
+    m_Layer.configure(&input, &output, blockSize);
+}
+
+void ClSpaceToDepthWorkload::Execute() const
+{
+    ARMNN_SCOPED_PROFILING_EVENT_CL("ClSpaceToDepthWorkload_Execute");
+    RunClFunction(m_Layer, CHECK_LOCATION());
+}
+
+arm_compute::Status ClSpaceToDepthWorkloadValidate(const TensorInfo& input,
+                                                   const TensorInfo& output,
+                                                   const SpaceToDepthDescriptor& desc)
+{
+    DataLayout dataLayout = desc.m_DataLayout;
+    const arm_compute::TensorInfo aclInputInfo = BuildArmComputeTensorInfo(input, dataLayout);
+
+    int32_t blockSize = boost::numeric_cast<int32_t>(desc.m_BlockSize);
+
+    const arm_compute::TensorInfo aclOutputInfo = BuildArmComputeTensorInfo(output, dataLayout);
+
+    const arm_compute::Status aclStatus = arm_compute::CLSpaceToDepthLayer::validate(&aclInputInfo,
+                                                                                     &aclOutputInfo,
+                                                                                     blockSize);
+    return aclStatus;
+}
+
+} //namespace armnn
diff --git a/src/backends/cl/workloads/ClSpaceToDepthWorkload.hpp b/src/backends/cl/workloads/ClSpaceToDepthWorkload.hpp
new file mode 100644
index 0000000..57ce5d4
--- /dev/null
+++ b/src/backends/cl/workloads/ClSpaceToDepthWorkload.hpp
@@ -0,0 +1,29 @@
+//
+// Copyright © 2017 Arm Ltd. All rights reserved.
+// SPDX-License-Identifier: MIT
+//
+
+#pragma once
+
+#include <armnn/Descriptors.hpp>
+
+#include <backendsCommon/Workload.hpp>
+#include <arm_compute/runtime/CL/functions/CLSpaceToDepthLayer.h>
+
+namespace armnn
+{
+arm_compute::Status ClSpaceToDepthWorkloadValidate(const TensorInfo& input,
+                                                   const TensorInfo& output,
+                                                   const SpaceToDepthDescriptor& desc);
+
+class ClSpaceToDepthWorkload : public BaseWorkload<SpaceToDepthQueueDescriptor>
+{
+public:
+    ClSpaceToDepthWorkload(const SpaceToDepthQueueDescriptor& descriptor, const WorkloadInfo& info);
+    void Execute() const override;
+
+private:
+    mutable arm_compute::CLSpaceToDepthLayer m_Layer;
+};
+
+} //namespace armnn
diff --git a/src/backends/cl/workloads/ClWorkloads.hpp b/src/backends/cl/workloads/ClWorkloads.hpp
index 6ba8138..256b68c 100644
--- a/src/backends/cl/workloads/ClWorkloads.hpp
+++ b/src/backends/cl/workloads/ClWorkloads.hpp
@@ -34,6 +34,7 @@
 #include "ClSoftmaxFloatWorkload.hpp"
 #include "ClSoftmaxUint8Workload.hpp"
 #include "ClSpaceToBatchNdWorkload.hpp"
+#include "ClSpaceToDepthWorkload.hpp"
 #include "ClSplitterWorkload.hpp"
 #include "ClStridedSliceWorkload.hpp"
 #include "ClSubtractionWorkload.hpp"