Deepwater turbidite lobe reservoirs have massive hydrocarbon potential and represent one of the most promising exploration targets for hydrocarbon industry. Key elements of turbidite lobes internal heterogeneity include the architectural hierarchy and complex amalgamations at each hierarchical level leading to the complex distribution of shale drapes. Due to limitation of data, to build models realistically honoring the reservoir architecture provides an effective way to reduce risk and improve hydrocarbon recovery. A variety of modeling techniques on turbidite lobes exist and can be broadly grouped into pixel-based, process-based, process-oriented, surface-based, object-based and a hybrid approach of two or more of these methods. The rationale and working process of methods is reviewed, along with their pros and cons. In terms of geological realism, object-based models can capture the most realistic architectures, including the multiple hierarchy and the amalgamations at different hierarchical levels. In terms of data conditioning, pixel-based and multiple-point statistics methods could honor the input data to the best degree. In practical, different methods should be adopted depending on the goal of the project. Such a review could improve the understanding of existing modeling methods on turbidite lobes and could benefit the hydrocarbon exploration activities of such reservoirs in offshore China.