Significance: In 2023, China marked a significant milestone in nuclear energy development, becoming the first nation to successfully commercialize fourth-generation high-temperature gas-cooled reactor (HTGR) technology. This accomplishment represents a substantial advancement in nuclear power generation, offering enhanced safety features, improved thermal efficiency, and greater operational flexibility compared with conventional reactor designs. Progress: This paper provides a comprehensive and systematic examination of recent advancements in coupled simulation methodologies for high-temperature pebble flow dynamics within pebble-bed reactor cores, focusing on elucidating the fundamental physical mechanisms governing granular flow behavior and heat transfer processes. These mechanisms are critical for optimizing HTGR performance metrics, operational reliability, and safety parameters. Historically, research in this domain has faced two primary limitations. First, there have been methodological shortcomings in accurately modeling the complex multiphysics phenomena involved. Second, there have been substantial computational resource requirements, virtually rendering full-core-scale numerical investigations of the HTR-PM reactor configuration impractical. This study makes significant breakthroughs by developing innovative computational frameworks and artificial intelligence-enhanced methodologies. It systematically examines three key research areas. The first area is modeling and simulation techniques, including the discrete element method (DEM) for pebble flow characterization, computation fluid dynamics—DEM coupling methodologies, and graphical processing unit (GPU)-accelerated DEM software development. The second is heating transfer mechanisms, encompassing intra-and inter-particle conduction models, convective heat transfer formulations, and high-temperature radiative heat transfer approaches. The third is artificial intelligence (AI)-enhanced methodologies featuring the use of neural networks for pebble residence time prediction, future state forecasting networks, and convolutional neural networks-based radiative view factor estimation. A significant component of this study is the creation of GRAPE, a proprietary three-dimensional DEM-based numerical model that incorporates GPU heterogeneous parallel acceleration. This approach facilitates efficient simulation of pebble flow dynamics with up to tens of millions of particles on standard workstations while maintaining superior computational accuracy. The study's particle-scale heat transfer model integrates neural network-accelerated radiative heat transfer computation with enhanced mesh search algorithms, demonstrating remarkable capability in predicting pebble-bed temperature distributions at reactor core scales. Furthermore, the introduction of two novel deep learning architectures represents a significant advancement in pebble flow research methodologies. RT-Net is a multi-branch convolutional network designed for real-time pebble flow residence time prediction, and Pre-Net is an image generation network employing guided learning principles to forecast pebble flow evolution. These AI-driven tools, integrated within our integrated computational framework, serve to not only address critical research gaps but also establish new paradigms for analyzing multiphysics coupling in high-temperature multiphase particle systems. This development marks a substantial advancement in HTGR core research and development through key innovations, including the GRAPE simulation platform, neural network-enhanced heat transfer modeling, and pioneering applications of deep learning in pebble flow prediction. Conclusions and Prospects: These advancements provide a theoretical foundation and the practical tools necessary to address the next generation of challenges in nuclear energy technology, particularly in reactor safety, operational optimization, and the development of even more advanced reactor designs. Through a review, elaboration, and synthesis of these key issues, this paper identifies critical challenges and future research directions for developing high-temperature gas-cooled reactor cores, as well as high-temperature multiphase particle flow and multiphysics coupling.