Background: Premature delivery interrupts the natural growth of the fetus. The postnatal healthy management of preterm infants still follows term standards after a postmenstrual age (PMA) of 40 weeks and there is a lack of research on the longitudinal dynamic postnatal growth tracks of preterm infants. Methods: Based on the database established by the Wuhan University Internet+ Early Childhood Development Alliance in China, information on preterm infants, including birth registration and health follow-ups from 2016 to 2022, was incorporated into the health management system. Standardized anthropometric measurements of preterm infants were recorded from birth to a corrected age (CA) of 36 months. A generalized additive model based on location, scale, and shape was used to establish the percentile values and growth curves. Results: In total, 79,514 preterm infants were included in this study, and the birth weights at each gestational age (GA) were similar to Chinese standards. When evaluated by term birth weight, we found that the proportions of extrauterine growth retardation at a PMA of 40 weeks were all above 10% in the GA <= 34-week groups and reached between 17.19% and 55.56% in very preterm infants (VPIs). There was a high incidence of preterm infants with a weight below the third percentile in VPIs when referring to term standards at CAs of 0, 6, 12, 24, and 36 months (p < 0.001). We established natural growth curves of the preterm population with different GAs between CAs of 0 and 36 months, which indicated that the weight/length of late preterm infants was close to term standards while the growth trajectory of VPIs consistently lagged behind (p < 0.001). Conclusion: Our study revealed the different growth trajectories of preterm infants with different GAs. A set of growth curves and percentile values for preterm infants of different GAs between CAs of 0 and 36 months were established, offering an optional method for growth assessment of this special population.