Abstract: This paper aimed at comparing appropriateness of three non-linear growth curves（Von Bertalanffy、Gompertz、Logistic）to describe growth curve of Longling yellow goats.The weight of 90 Longling yellow goats（45 male and 45 female） at birth, 3, 6, 9, 12,18,24 and 36 months of age were measured in this experiment. Fitting degree（R~2）, mean square error（MSE） and mean absolute error（MAE） were used as evaluation indexes of the three growth curve models, and Pearson correlation coefficient between fitting estimated value and measured value was used as validation index to analyze the model that best fit the growth curve of Longling yellow goats. The results showed that the inflection age of male and female Longling yellow goats estimated by Von Bertalanffy model were 3.96 months, 4.53 months, respectively, and the inflection weight were 13.12 kg, 12.33 kg respectively, and the female goats reached the mature weight earlier than the male goats. Von Bertalanffy model was the best fitting method for the growth curves of male and female goats, with the maximum R~2 value（male and female goats were 0.988 3, 0.986 9, respectively）, the minimum MSE value（male and female goats were 3.176 7, 3.419 8, respectively） and MAE value（male and female goats were 0.153 8, 1.290 0, respectively）. Among the three models, the Pearson correlation coefficient between the estimated value and the measured value of body weight in male and female goats was the largest in Von Bertalanffy model, which were 0.994 2, 0.993 5 for male and female goats, respectively. The results indicated that Von Bertalanffy model had the best fitting effect on the growth curve of Longling yellow goats.