Objective To explore the effect of internet of things-based power bicycle training or quadriceps training alone on pain and quality of life in patients with knee osteoarthritis in a community setting. Methods Patients with knee osteoarthritis who were admitted to West China Hospital of Sichuan University between April and July 2022 were selected. They were randomly divided into a power bicycle training group, a quadriceps muscle training group and a control group by random number table method. The primary outcome was improvement in knee pain, assessed by the Numerical Rating Scale (NRS) score. The secondary outcome was health-related quality of life, assessed by the 36-Item Short Form Health Survey (SF-36) score. Outcomes were assessed at baseline and 4, 8, and 12 weeks after the initial intervention. The statistical analysis was conducted using generalized estimating equations. Results A total of 72 patients were included, with 24 in each group. There was no significant difference in age, gender or other demographic characteristics among the three groups (P>0.05). The results of generalized estimating equations showed that there were interaction effects (group × time) on the NRS score, SF-36 physical functioning score, SF-36 bodily pain score, and SF-36 vitality score (P<0.05), while there was no interaction effect (group × time) on the role physical score, general health score, social functioning score, role emotional score or mental health score of SF-36 (P>0.05). At baseline, there was no statistically significant difference in the NRS score or SF-36 scores among the three groups (P>0.05). After 12 weeks of intervention, the two training groups were better than the control group in the NRS score, SF-36 physical functioning score, and SF-36 bodily pain score, the power bicycle training group was better than the quadriceps training group in the NRS score, the power bicycle training group was better than the control group in the SF-36 social functioning score, and the differences were statistically significant (P<0.05). Conclusions In a community setting, 12 weeks of internet of things-based power bicycle training and quadriceps training can significantly improve joint pain, physiological function and physical pain indicators in patients with knee osteoarthritis, and the power bicycle training is better than the quadriceps training in improving the knee pain of patients.