Motivated by challenges of AI adoption and inspired by recent developments in Computer Vision and Natural Language Processing, this paper proposes a Foundation Model for inventory management. Built on a unified model architecture and a streamlined training process, our model exhibits exceptional versatility and scalability. It can deal with thousands of heterogeneous products, exploit cross-learning opportunities in a large product portfolio, and has zero-shot learning capabilities that make it suitable for managing new products with very limited historical data is available. Based on a real-world retail dataset we show that these benefits can be achieved without sacrificing performance: Our Foundation Model consistently outperforms various state-of-the-art models in terms of inventory distortion costs. Our contributions are relevant for the Information Systems community, highlighting how AI-based decision-making can be effectively implemented in complex business environments connecting cutting-edge developments in Operations Management and Machine Learning to practical applications creating real, tangible economic value through Business Analytics.