Apricot (Prunus armeniaca L.) is a world-wide highly appreciated fruit, with its attractive colour, soft texture and typical flavour. In the current investigation, 68 apricot accessions collected from Jammu and Kashmir, India were analysed to determine the measure of variation using 12 qualitative and 16 quantitative traits with an aim to identify superior apricot accessions with excellent fruit quality traits. High phenotypic variability was observed among the studied apricot accessions with significant differences among most of the qualitative and quantitative traits. Fruit-related characteristics including fruit weight, fruit length, fruit ratio, fruit firmness, fruit shape, fruit suture, stone weight, kernel weight and stone shape were the most diverse with a high coefficient of variation (>30%). One way analysis of variance showed significant differences (P < 0.0001) among all the quantitative traits. Significant positive and negative correlations were observed between all the agronomically important fruit quality traits. The principal component analysis (PCA) revealed that 75.34% of the variability was defined by the first eight components. The unweighted pair group method with arithmetic mean (UPGMA) dendrogram, based on all traits measured grouped the accessions into two main clusters with several sub-clusters. Both UPGMA dendrogram and PCA scatter plot formed a cluster of 14 accessions, having the highest values regarding most important fruit quality traits such as fruit weight, fruit length and width, fruit ratio, stone weight and kernel weight can be treated as potentially superior accessions. These accessions can be used directly for cultivation and in future apricot breeding programmes. The present findings are promising for genetic resource management, cultivar improvement and commercial applications of apricot in Jammu and Kashmir, India.