激光增材制造高熵合金因其独特的跨尺度不均匀组织及工程应用潜力而备受关注，但其疲劳性能仍是制约工程化应用的关键瓶颈。 基于此，本文系统梳理了激光增材制造高熵合金的疲劳行为研究进展。 现有研究表明，其疲劳失效机制呈现出由缺陷主导向组织主导的转变趋势。 其中，在疲劳早期阶段，缺陷尺寸、形状及其连通性起主导作用；而在疲劳裂纹萌生后，晶格旋转、形变孪晶、内应力分布以及析出相等因素进一步决定裂纹扩展行为及其速率。 进一步对不同合金体系的对比分析表明，析出强化型高熵合金通常表现出更高的疲劳抗力，因而有望成为提升疲劳性能的重要研究方向。 基于此，本文提出后续研究应融合缺陷控制与组织调控，建立面向激光增材制造的疲劳预测与评估体系，并以应用需求为牵引，推动高熵合金工程化转化与应用。

Laser additive manufactured high-entropy alloys (HEAs) have emerged as a class of structurally promising materials owing to their intrinsic engineering potential and the formation of heterogeneous microstructures across multiple length scales. Despite these advantages, the fatigue performance remains a major barrier to their reliable structural application. In this review, recent progress in understanding the fatigue behavior of laser additive manufactured HEAs was systematically assessed. Current findings indicate that the governing fatigue failure mechanism evolves from a defect-dominated regime to a microstructure-mediated regime. At the early stage of fatigue damage, premature failure is primarily controlled by the size, morphology, and interconnectivity of process-induced defects. Once cracks are initiated, their subsequent growth is further modulated by microstructural features, including lattice rotation, deformation twinning, the partitioning of high- and low-strain regions, and precipitate interfaces, all of which contribute to variations in crack propagation kinetics. Cross-system comparisons further reveal that precipitation-strengthened HEAs generally exhibit superior fatigue resistance, highlighting their particular promise for future alloy design. In light of the current limitations, future research should move beyond isolated defect mitigation or microstructural optimization and instead integrate both aspects into a unified framework for fatigue prediction and performance evaluation tailored to laser additive manufacturing. Such an approach is expected to accelerate the engineering translation of HEAs and facilitate application-driven demonstration and validation under service-relevant conditions.