Abstract: Global cloud and precipitation data derived from satellite-based passive microwave sensors are essential for understanding the formation and evolution mechanisms of hazardous weather events and climate change. Focusing on Micro-Wave Radiation Imager-II (MWRI-II) onboard FY-3F, China’s new-generation morning-orbit operational meteorological satellite, a retrieval algorithm based on Bayesian theory and a highly representative and physically accurate prior database is developed. The algorithm retrieves surface precipitation rates and vertical profiles of multiple hydrometeors, including cloud liquid water, rainwater, and ice water, and is further validated and evaluated. The evaluation results based on simulations demonstrate that the proposed algorithm has very high accuracy, with the mean bias of the retrieved physical quantities and reference values of any hydrometeor being within 10 g·m⁻². The comparison of the retrieval results based on FY-3F measurements with FY-3G Multiple-Hydrometeor Profile (MHP) products and Goddard Profiling Algorithm (GPROF) products of Global Precipitation Measurement Mission (GPM) shows that for the statistical analysis over China and its surrounding regions during May–July 2024, the spatial distributions of the three datasets are generally consistent with correlation coefficients above 0.60, reaching up to 0.83. In addition, the relative biases between the retrieved parameters and those in the MHP product are all about 10%. Note that the algorithm does not rely on any other auxiliary data and can provide technical support for the operational deployment of the cloud-precipitation products for FY-3F. Furthermore, the algorithm can also be applied to similar passive microwave radiometers with similar channel configurations onboard other Fengyun satellites, offering significant potential for future multi-satellite collaborative observation strategies.