<p>Artificial Intelligence (AI) is rapidly reshaping agricultural systems by enabling data-driven decision-making, predictive analytics, and automated farm management. This review critically synthesizes recent advances in computational models and algorithms applied to sustainable and climate-smart agriculture. It examines the role of machine learning techniques (e.g. Random Forest, Support Vector Machines, XGBoost), deep learning architectures (e.g. Convolutional Neural Networks, recurrent models, transformer-based frameworks), and computer vision systems for crop monitoring, disease detection, yield prediction, livestock management, and precision irrigation. The review further analyzes the integration of Internet of Things (IoT) sensor networks with AI models for real-time data acquisition and adaptive control. Emerging approaches, including generative AI and large language models, are evaluated for their potential to provide context-aware advisory systems and decision support. Across applications, AI demonstrates measurable improvements in prediction accuracy, input optimization, early stress detection, and climate risk forecasting. However, challenges related to model generalization, data heterogeneity, computational scalability, and deployment in resource-constrained environments remain significant barriers. The study identifies immediate research priorities in robust model design and explainable AI, as well as long-term directions toward autonomous, self-learning agricultural ecosystems. Overall, this review highlights how advanced computational methods are driving the transition toward intelligent, resilient, and sustainable agricultural systems.</p>