The large spatial variability in monsoon rainfall over India demands for regional models for predicting the seasonal rainfall. Hence, models were developed for predicting seasonal (June-September) rainfall of three regions (north, middle and south) of Gujarat using multiple regression technique. The monthly weather data of 30 years of Anand (1980-2009), 22 years (1987-2009) of Navsari and 27 years (1983-2009) of SK Nagar were used. The models were validated with independent data set of four year (2006-2009). The best models were selected based on higher R2 and lower model error. Four models were obtained; 2 for Anand (middle Gujarat) and one each for SK Nagar (north Gujarat) and Navsari (south Gujarat). Different models explained 74 to 93% variability in seasonal rainfall with models error ranging between -2.5 to 5.1%. During the validation period the performance of model was quite satisfactory with model error rainging between -12.6 to 2.6%. All the models were used to predict the rainfall for 2010 season. Results suggested that the rainfall would be higher than the normal rainfall in all the three regions. Navsari (south Gujarat) is expected to receive 1529.0 mm (14.5% higher than the normal), Anand (middle Gujarat) is expected to get 1294.0 to 1363.0 mm (62-71% higher) and SK Nagar (north Gujarat) is expected to receive 770.0 mm (40% above normal) rainfall during June to September period of year 2010.