In this talk, we’ll discuss emerging trends in generative AI for edge devices and the key role of transformer-based neural networks. We’ll review the distinct attributes of transformers, their advantages over conventional convolutional neural networks and how they enable generative AI. Then we will cover key requirements that must be met for neural processing units (NPU) to support transformers and generative AI in edge device applications. We will use transformer-based generative AI examples to illustrate the efficient mapping of these workloads onto the enhanced Synopsys ARC NPX NPU IP family.