Developed and integrated the Bailing Moe model architecture within the ml-explore/mlx-swift-examples repository, focusing on enhancing language-task performance through new configurations and model definitions. Leveraged expertise in machine learning, model architecture design, and Swift development to deliver an end-to-end feature that emphasized configurability and code quality. The work involved designing modular components to support rapid experimentation and downstream usage, ensuring stability and readiness for further development. No major bugs were reported during the period, reflecting careful implementation and testing. This contribution provided a robust foundation for future machine learning experiments in a Swift-based example environment.