This study was conducted to establish reliable prediction models for net energy (NE) values of rapeseed meals (RSM) and cottonseed meals (CSM) based on chemical composition and apparent metabolic energy (AME) for Avian broilers aged from 1 to 21 days, and to compare the predictive ability of modeling the two samples (RSM and CSM) separately and together. NE values of RSM and CSM were measured as the sum of NE values for maintenance (NEm) and for production (NEp), and 15 kinds of RSM and 15 kinds of CSM were measured. The NEm was measured by regression method with 4 feeding levels including ad libitum feeding and restricted feeding by 30%, 50% and 70%, respectively. NEp was measured by the method of substitution. Seven-day-old Avian broilers with an average weight of (97.3±4.0) g were randomly allotted into every feeding level and every sample of rapeseed meal and cottonseed meal with 6 replicates each and 2 chickens in each replicate. Proximate chemical composition of RSM and CSM was measured. The linear regression analysis was carried out between NE values, AME values and chemical composition based on the two samples separately and together. The results showed as follows: the NE values of RSM and CSM for broilers aged from 1 to 21 days were from 4.72 to 7.22 MJ/kg DM and from 4.73 to 7.08 MJ/kg DM, respectively; the R2 of the optimum regression equations for the two samples based on AME combined with chemical composition were 0.995 and 0.998, respectively; the relative standard deviations (RSD) were 0.052 and 0.033 MJ/kg DM, respectively; the R2 of the optimum regression equation for the two samples together based on AME combined with chemical composition was 0.995, and the RSD was 0.052 MJ/kg DM. The R2 of the optimum regression equations for the two samples based on the chemical composition were 0.973 and 0.985, respectively; the RSD were 0.123 and 0.100 MJ/kg DM, respectively; the R2 of the optimum regression equation for the two samples together based on the chemical composition was 0.973, and the RSD was 0.123 MJ/kg DM. The results indicate that the NE values of RSM and CSM are accurate and the reliable regression equations based on AME combined with chemical composition are better than the reliable regression equations only based on chemical composition; both of the prediction models for NE values of the two samples separately and together are reliable.［Chinese Journal of Animal Nutrition, 2011, 23（10）：1769 -1774］