Abstract: With the rapid advancement of artificial intelligence technology, the demand for computing power has experienced exponential growth. Traditional computing architectures struggle to meet the ever-increasing demand, making the development of XPU a core driving force in modern intelligent computing. As the power consumption and current requirements of high-performance computing chips surge dramatically, power supply technologies for high computing XPUs face formidable challenges, necessitating more efficient and reliable power solutions to support the large-scale computational demands. Regarding the development status and existing problems of high computing XPU power supply technologies, a comprehensive review of critical power supply technologies for high computing XPUs is provided, encompassing the evolution of power architectures, high-ratio DCX topology technologies, high-dynamic control technology, and high-density integration solutions for XPU power systems. Furthermore, the technical challenges faced by XPU power supply technologies in architecture design, topology optimization, power device selection, dynamic control methodologies, and implementation approaches are systematically analyzed. Finally, future development directions in this field are outlined, addressing potential solutions and emerging trends, which can provide a reference for the subsequent research.