Recently, quantitative products of space-borne remote sensing have gradually become important inputs for various applications in China, widely serving fields such as meteorology, oceanography, natural resources, ecological environment, and Earth system research. Although the rapid growth in the number of remote sensing satellites, the widespread use of remote sensing data is severely restricted due to their differences in data formats, processing methods, and application fields of different satellite systems. To reduce the threshold for using remote sensing data, we proposed a standardized data product model based on Data Square (DS) according to the remote sensing information model and data engineering model. The multi-source spatio-temporal data are normalized and organized under a unified geographic grid framework from the perspectives of data product levels, observation methods, and product attributes. Four types of control samples, namely geometric, radiometric, classification, and parametric, are applied for the relative truth constraints of data squares. Synthetic product production is carried out for data square collections acquired by different constellations, satellites, and remote sensors at different times and locations, forming spatio-temporally consistent remote sensing standard quantitative products and file codes. The advantages and disadvantages of the proposed models are further investigated for the existing Analysis Ready Data (ARD) models in the space-borne remote sensing, providing technical support for the vision of universal application across multiple satellites.