Abstract: Carbonate reservoirs are a major area for onshore oil and gas exploration in China. A number of large and medium- sized oil and gas fields have been discovered in carbonate reservoirs in the Tarim, Sichuan, Ordos and other basins. With continuous exploration, deep and ultra-deep carbonate reservoirs with strong heterogeneity will play an important role in large-scale exploration. However, there are some issues with the related seismic prediction technologies, such as weak theoretical methods and low prediction accuracy. During the 13th Five Year Plan period, important progress has been made in theoretical research and new technologies. Among them, innovative methods, such as a petrophysical model for forward and inversion modeling of complex wave fields of fracture-porous media, petrophysical analysis and reservoir pore structure identification by digital carbonate rock core data, have laid an important foundation for the development of new technologies for reservoir prediction. New technologies, such as small faults identification by seismic gradient structure tensor, quantitative prediction of fracture- cavity reservoirs by cloud transform stochastic simulation, and gas reservoir detection based on pre-stack elastic parameter inversion and frequency division attributes, have effectively improved the recognition accuracy of fault, reservoir and fluid. On this basis, combined with the research status of seismic prediction technologies for carbonate reservoirs with strong heterogeneity, some suggestions for technical development are proposed. According to the development trend of “deep fusion, refinement and intelligentization”, basic theoretical research is strengthened, such as petrophysical modeling methods based on fracture-porous media with heterogeneity of pore shape, fracture-induced anisotropy, and characteristics of dispersion and attenuation. New technologies should be focused on, such as refined seismic prediction technology for reservoir sensitive attributes based on two-phase medium frequency, wave dynamic characteristics such as dispersion and attenuation, seismic prediction technology for reservoir pore structure based on petrophysical analysis of digital core, and quantitative prediction and fluid detection of carbonate reservoirs by artificial intelligence.