Osteoarthritis (OA) is a chronic, progressive and degenerative disease, and its incidence is increasing on a yearly basis. However, the pathological mechanism of OA at each stage is still unclear. The present study aimed to explore the underlying mechanism of dihydroartemisinin (DHA) in terms of its ability to inhibit osteoclast activation, and to determine its effects on OA in rats. Bone marrow-derived macrophages were isolated as osteoclast precursors. In the presence or absence of DHA, osteoclast formation was assessed by tartrate-resistant acid phosphatase (TRAP) staining, cell viability was assessed by Cell Counting Kit-8 assay, the presence of F-actin rings was assessed by immunofluorescence, bone resorption was determined by bone slices, luciferase activities of NF-κB and nuclear factor of activated T cell cytoplasmic 1 (NFATc1) were determined using luciferase assay kits, the protein levels of biomolecules associated with the NF-κB, MAPK and NFATc1 signaling pathways were determined using western blotting, and the expression of genes involved in osteoclastogenesis were measured using reverse transcription-quantitative PCR. A knee OA rat model was designed by destabilizing the medial meniscus (DMM). A total of 36 rats were assigned to three groups, namely the sham-operated, DMM + vehicle and DMM + DHA groups, and the rats were administered DHA or DMSO. At 4 and 8 weeks postoperatively, the micro-architecture of the subchondral bone was analyzed using micro-CT, the thickness of the cartilage layers was calculated using H&E staining, the extent of cartilage degeneration was scored using Safranin O-Fast Green staining, TRAP-stained osteoclasts were counted, and the levels of receptor activator of NF-κB ligand (RANKL), C-X-C-motif chemokine ligand 12 (CXCL12) and NFATc1 were measured using immunohistochemistry. DHA was found to inhibit osteoclast formation without cytotoxicity, and furthermore, it did not affect bone formation. In addition, DHA suppressed the expression levels of NF-κB, MAPK, NFATc1 and genes involved in osteoclastogenesis. Progressive cartilage loss was observed at 8 weeks postoperatively. Subchondral bone remodeling was found to be dominated by bone resorption accompanied by increases in the levels of RANKL, CXCL12 and NFATc1 during the first 4 weeks. DHA was found to delay OA progression by inhibiting osteoclast formation and bone resorption during the early phase of OA. Taken together, the results of the present study demonstrated that the mechanism through which DHA could inhibit osteoclast activation may be associated with the NF-κB, MAPK and NFATc1 signaling pathways, thereby indicating a potential novel strategy for OA treatment.