Abstract: Flotation is a core technology for mineral separation in mineral processing, and the analysis of its mechanism is crucial for optimizing the process and developing efficient reagents. However, traditional quantum chemical calculations are limited by high costs and model scale, making it difficult to simulate multi−component actual flotation systems and accurately describe the dynamic interactions at the interface between minerals, reagents, and water, thus restricting the depth of mechanism research. The collaborative application of quantum chemistry and machine learning retains the high−precision description of interatomic interactions from the former while leveraging the latter to enhance computational efficiency, opening up a new path for flotation mechanism research. This paper reviews the application progress of quantum chemistry in the field of mineral flotation, presenting research advancements in the study of mineral surface geometry and electronic properties, reagent design, and the analysis of the mechanism of reagent−mineral interface interactions. It also highlights the advantages of quantum chemistry−driven machine learning methods in high−throughput screening of reagents, which can significantly shorten the screening cycle and efficiently explore the structure−activity relationship of reagents. The paper further looks forward to the application prospects of this method in the calculation of complex interface reaction systems. At the same time, it points out the challenges currently faced in large−scale system computational efficiency, water layer simulation, and multi−parameter simulation under temperature changes. In the future, it is expected that by combining quantum chemistry calculations with machine learning, models that can simultaneously reflect atomic−level interactions and the overall flotation process can be constructed, thereby more accurately describing the dynamic process of reagent−mineral interactions.