The soluble solid content (SSC), one of the important index for evaluating the qualities, in a single grape were measure real-timely and non-destructively in this study, This might help to improve the management of the fruits as well as prolong their storage time, satisfying the needs of different customers. The near infrared (NIR) spectra of grapes was detected by using a hand-held NIR spectrometer, wavelength ranging from 950 to 1, 650 nm. Based on partial least squares (PLS) data, the prediction model for SSC content in grapes was established. In order to reduce redundancy and uninformative variables while increase prediction accuracy and stability of this model, sensitive wavelength variables were selected by uninformative variable elimination (UVE) and random frog (RF), respectively. The results showed that the SSC predictive model established by RF-PLS is better than the ones done by full-spectrum PLS and UVE. R2c, R2_cv and R2pof RF-PLS were 0.960 5, 0.933 4 and 0.930 4, and RMSEC, RMSECV and RMSEP of it were 0.638 2, 0.829 9 and 0.868 8, respectively. Our results showed that the spectra based on portable NIR spectrometer could be successfully applied in prediction of SSC content in a single grape with high prediction accuracy after wavelength selection for the model.