Abstract: In order to improve the ability of predicting the yields of greenhouse cucumber, a model for fruit growth nf greenhouse cucumber was developed based on the effects of temperature and radiation on fruit growth using production of thermal and PAR (TEP) as prediction index. Experiment with other planting data was carried out in greenhouse to collect and validate the model. The results showed that the simulated values agreed well with the measured ones. The determined coefficient (R2) and root mean squared error (RMSE) between the simulated and measured values of fruit length on different leaf axils was 0.7325 and 1.64 cm, respectively. The determined coefficient (r2) and root mean squared error (RMSE) between the simulated and measured values fruit diameter on different leaf axils was 0.5885 and 0.40 cm, respectively. The determined coefficient (R2) and root mean squared error (RMSE) between the simulated and measured values of fruit fresh weight on different leaf axils was 25.04 g and 0.6782 respectively. And the determined coefficient (R2) and root mean squared error (RMSE) between the model based on growing degree days (GDD) simulated and measured values of fruit fresh weight on different leaf axils was 0.6068 and 28.52 g. The prediction accuracy of TEP based model is 12.21% higher than that of GDD. Based on the results it can be concluded that the model can give a satisfactory prediction of fruit fresh weight on different axils. The model developed in this study is applicable and can provide theoretical basis and decision support for the production of greenhouse cucumber.