diff --git a/src/backends/cl/workloads/ClSoftmaxWorkload.cpp b/src/backends/cl/workloads/ClSoftmaxWorkload.cpp
new file mode 100644
index 0000000..cbca766
--- /dev/null
+++ b/src/backends/cl/workloads/ClSoftmaxWorkload.cpp
@@ -0,0 +1,48 @@
+//
+// Copyright © 2020 Arm Ltd. All rights reserved.
+// SPDX-License-Identifier: MIT
+//
+
+#include "ClSoftmaxWorkload.hpp"
+#include "ClWorkloadUtils.hpp"
+
+#include <aclCommon/ArmComputeTensorUtils.hpp>
+#include <aclCommon/ArmComputeUtils.hpp>
+
+#include <cl/ClTensorHandle.hpp>
+
+namespace armnn
+{
+
+arm_compute::Status ClSoftmaxWorkloadValidate(const TensorInfo& input,
+                                              const TensorInfo& output,
+                                              const SoftmaxDescriptor& descriptor)
+{
+    const arm_compute::TensorInfo aclInputInfo = armcomputetensorutils::BuildArmComputeTensorInfo(input);
+    const arm_compute::TensorInfo aclOutputInfo = armcomputetensorutils::BuildArmComputeTensorInfo(output);
+
+    unsigned int aclAxis = ComputeSoftmaxAclAxis<unsigned int>(descriptor, input);
+    return arm_compute::CLSoftmaxLayer::validate(&aclInputInfo, &aclOutputInfo, descriptor.m_Beta, aclAxis);
+}
+
+ClSoftmaxWorkload::ClSoftmaxWorkload(const SoftmaxQueueDescriptor& descriptor, const WorkloadInfo& info,
+                                     std::shared_ptr<arm_compute::MemoryManagerOnDemand>& memoryManager)
+        : BaseWorkload<SoftmaxQueueDescriptor>(descriptor, info)
+        , m_SoftmaxLayer(memoryManager)
+{
+    m_Data.ValidateInputsOutputs("ClSoftmaxWorkload", 1, 1);
+
+    arm_compute::ICLTensor& input  = static_cast<ClTensorHandle*>(m_Data.m_Inputs[0])->GetTensor();
+    arm_compute::ICLTensor& output = static_cast<ClTensorHandle*>(m_Data.m_Outputs[0])->GetTensor();
+
+    unsigned int aclAxis = ComputeSoftmaxAclAxis<unsigned int>(m_Data.m_Parameters, info.m_InputTensorInfos[0]);
+    m_SoftmaxLayer.configure(&input, &output, m_Data.m_Parameters.m_Beta, aclAxis);
+}
+
+void ClSoftmaxWorkload::Execute() const
+{
+    ARMNN_SCOPED_PROFILING_EVENT_CL("ClSoftmaxWorkload_Execute");
+    RunClFunction(m_SoftmaxLayer, CHECK_LOCATION());
+}
+
+} // namespace armnn