The giant panda (Ailuropoda melanoleuca), as an endemic species of China, is also a flagship species of wildlife Conservation. However, the accuracy of individual identification of giant pandas is typically problematic due to the high error rate of microsatellite typing methods. In this study, a panel of 61 SNP makers was developed for individual identification. Of these, 55 polymorphic SNP markers were successfully genotyped. The observed heterozygosity (Ho), expected heterozygosity (He) and polymorphic information content (PIC) ranged from 0.056 to 0.611 (mean 0.352), 0.131–0.509 (mean 0.389) and 0.099–0.389(mean 0.305), respectively. No SNPs significantly deviated from Hardy–Weinberg equilibrium (P < 0.05). Our study provides a framework for improving the accuracy of individual identification of giant pandas, especially for poor quality samples collected from the wild.