Abstract: The apparent amylose content (AAC) is one of the important parameters to affect the cooking and processing characteristics. The chemical measurement of rice AAC is very expensive and time consuming. In recent years, near infrared spectroscopy(NIRS), a rapid analytical technique is used for testing agricultural products. The prediction models for rice samples with different particles and types were established by NIRS. The correlation coefficient between predicted value and chemical value was 0.95, standard error of prediction (SEP) was 0.56, and mean relative error (MER) was 3.1%.