Soluble solids content (SSC) and firmness index (FI) are key factors that affect the taste of strawberries. Accordingly, a general predictive model was established for strawberry quality parameters, SSC and FI, based on common features. First, spectral features were extracted using competitive adaptive reweighted sampling (CARS), the successive projection algorithm (SPA), and uniformative variable elimination (UVE). Subsequently, decision models were constructed using partial least squares regression (PLSR), extreme learning machine (ELM), and least squares support vector machine (LS-SVM). The whale optimization algorithm (WOA) was used to optimize the parameters of the LS-SVM model to ultimately establish a general predictive model based on the common features of SSC and FI. The results indicated that spectral noise can be effectively reduced using the Savitzky-Golay (SG) convolution smoothing technique for preprocessing. The CARS-LS-SVM model exhibited the best single-indicator predictive performance for SSC and FI, with correlation coefficients of 0.937 and 0.898 and residual predictive deviations of 2.87 and 2.28, respectively. The SSC and FI features extracted using the UVE method had the highest overlap rate. The LS-SVM dual-indicator model that was established based on common features achieved effective predictions of SSC and FI, with correlation coefficients of 0.922 and 0.871 and residual predictive deviations of 2.58 and 2.04, respectively. Near-infrared spectroscopy could simultaneously predict the SSC and FI of strawberries. The results of this study can serve as a theoretical reference for the development of general predictive models capable of determining multiple strawberry quality parameters.