Flexible electronic devices, renowned for their exceptionally thin, lightweight, adaptable, customizable, and cost-effective nature, are revolutionizing industries such as wearable healthcare, implantable neural recording, human-machine interaction, and the Internet of Things. In recent times, the advent of stable and high-yield thin-film transistor (TFT) technology has paved the way for the fabrication of large-scale integrated circuits on flexible substrates. Simultaneously, the fusion of flexible multimodal sensing with localized collaborative processing, coupled with the emergence of high-performance edge intelligence, has ignited a demand for flexible processors adept at efficiently accelerating neural network inference. Nevertheless, traditional hardware architectures have shown considerable constraints in terms of energy efficiency, reliability, and mechanical adaptability.