This report presents a description of five open-source repositories developed as part of CENTRIC Work Package 3 (WP3)’s activities and published as CENTRIC’s AI-based MIMO toolset in deliverable D3.4. The repositories include software implementation and documentation of simulation environments and AI-based MIMO algorithms developed in WP3 as part of the physical layer processing component of the AI native Air-Interface (AI-AI) concept which CENTRIC is developing. Each repository focuses on specific MIMO-AI algorithm such as reinforcement learning for beam management in Integrated Sensing and Communication (ISAC) scenarios, AI techniques for wide to narrow beam prediction, multi-user MIMO neural network-based receiver, transfer learning techniques for neural receiver and learning based beam alignment. This report provides detailed description of the problem addressed, system model and simulation environment, the developed AI algorithm, and usage example and 
results obtained from each of the repositories. These repositories offer a comprehensive set of MIMO-AI solutions allowing reproducibility of CENTRIC’s research output while serving as the basis for further development of AI-AI physical layer methods.