Objective: Portable near-infrared (NIR) spectroscopy and stoichiometric methods were used to predict the spoilage time of yellow peach. Methods: Diffuse reflectance spectra of yellow peach samples were collected using a portable NIR spectrometer. Spectral preprocessing techniques were employed to enhance data features, and partial least squares (PLS) regression was applied to establish a prediction model upon the spoilage time and NIR data of yellow peach. The model performance was evaluated by both root mean square error (RMSE) and coefficient of determination (R²). Results: An R² value of 0.63 and RMSE of 4.09 days were achieved for the spoilage time prediction for yellow peach. Conclusion: NIR spectroscopy combined with chemometric methods can provide a non-destructive and accurate prediction of yellow peach spoilage time.