A tool to effectively address the runtime management of AI applications component placement and resource selection in the Computing Continuum. Through a Random Search combined with a Stochastic Local Search algorithm, SPACE4AI-R copes with runtime workload fluctuations by identifying the cost-optimal reconfiguration of the initial production deployment, while providing performance guarantees across heterogeneous resources including Edge devices and servers, Cloud GPU-based Virtual Machines and Function as a Service solutions. Experimental results show that our tool efficiently finds placement reconfigurations in a real use case of identifying wind turbines blade damage, and can manage large-scale systems providing remarkable cost savings over static placements, while keeping execution time in the order of seconds.