<p>Hyacinth (<i>Hyacinthus orientalis</i> L.) is a popular floricultural crop. Its cultivars exhibit a wide range of phenotypic variations, especially flower colours. Yet, the cultivar pedigree was poorly recorded, impeding efficient breeding in producing novel cultivars. In addition, scarce genomic resource of the species hinders the exploration on the molecular mechanism in controlling the diversification of floral colour. In this study, transcriptome profiling was conducted on seven hyacinth cultivars, each representing a major flower colour. RNA-Seq libraries were prepared from 189 samples that were collected in three perianth partitions at three consecutive developmental stages in biological triplicates. A total of 1,256.8 gigabytes data were generated. The reproducibility and variability of our dataset were assessed through correlation analysis and principal component analysis, respectively. In addition, the usability of the dataset was demonstrated by the identification of differentially expressed genes, functional annotation and functional enrichment analysis. This study provides the first spatiotemporal profiling of the gene expression of hyacinths, contributing to molecular breeding of hyacinth cultivars with novel flower colours.</p>