OBJECTIVE: Transcutaneous auricular vagus nerve stimulation () may be an innovative treatment for symptoms of knee osteoarthritis (OA) due to possible shared pathological mechanisms between diminished parasympathetic function, central pain mechanisms, and knee pain. Thus, we sought to test the safety and preliminary efficacy of tVNS in people with knee OA. DESIGN: A pilot trial in which participants received a 60-min tVNS was conducted. At baseline, immediately after, and 15 ​min after tVNS, we assessed knee pain, pressure pain threshold (), temporal summation (), conditioned pain modulation (), and high-frequency power of heart rate variability (). We examined the extent to which these outcome measures changed after tVNS using linear mixed models. RESULTS: 30 participants with knee OA were included, and all completed the intervention without any major side effects. Compared to baseline, knee pain was reduced by 1.27 (95 ​% CI, -1.74, -0.80) immediately after and by 1.87 (-2.33, -1.40) 15 ​min after tVNS; CPM improved by 0.11 (0.04, 0.19) and 0.07 (-0.01, 0.15); and HF improved by 213.29 (-0.38, 426.96) and 234.17 (20.49, 447.84). PPT and TS were not changed after tVNS. CONCLUSIONS: Our preliminary data demonstrated that tVNS may be a safe pain-relieving treatment for people with knee OA. Our findings suggest that improvement of knee pain might be derived from improvement of parasympathetic function and central pain mechanisms as no local therapy was applied. A large study is needed to confirm that tVNS is a novel intervention to ameliorate knee pain in people with knee OA. CLINICAL TRIAL: ClinicalTrials.gov (NCT05625178).