Dendrobium officinale Kimura et Migo is a kind of herb with high medicinal, ornamental, and commercial value, and is rich in polysaccharides. Polyploid breeding is an important breeding method for the genome doubling of medicinal species to increase biomass and polysaccharide production. Previous studies have revealed comparative transcriptome analysis and polysaccharide biosynthesis across the growth stages and plant parts, but there have been no studies dissecting such genes and pathways in tetraploid D. officinale. Therefore, this study aimed to unravel the molecular mechanisms of the increase in polysaccharide content in tetraploid D. officinale via the generation of four transcriptomic libraries for protocorm-like bodies and six-month-old seedlings of both diploid and tetraploid D. officinale plants. In this study, a total of 230,786,618 clean reads remained with a total of 34.62 Gb nucleotides generated; 274,403 unigenes were assembled, of which 73.99% were annotated to at least one of the protein databases; and of 17,451 unigenes, 6.35% were annotated to all seven protein databases (NR, NT, KO, Swiss-Prot, FAM, GO, and KOG). Putative genes encoding enzymes related to polysaccharide biosynthetic pathways were determined. RT-qPCR for 11 randomly selected genes involved in polysaccharides indicated consistency with RNA-Seq data and polysaccharide content. The expressions of nine genes were higher in tetraploid than in diploid plants, while the expressions of the other two genes encoding bifunctional enzymes were the opposite. This study has provided a foundation for subsequent works regarding the biosynthetic pathways of metabolites involved in the autoploidy of Dendrobium species in general, and D. officinale in particular.