OBJECTIVE: The main clinical manifestations of knee osteoarthritis are pain, swelling, stiffness, and limited activity, which have a serious impact on the life of patients. Exercise therapy can effectively improve the related symptoms of patients with knee osteoarthritis. This paper uses the method of network meta-analysis to compare the efficacy of different exercise types in the treatment of knee osteoarthritis. METHODS: CNKI, WanFang, PubMed, Embase, Cochrane Library, Web of Science, Scopus, Ebsco, SinoMed, and UpToDate were searched with Chinese search terms “knee osteoarthritis, exercise therapy” and English search terms “knee osteoarthritis, exercise”. Randomized controlled trials on the application of different exercise types in patients with knee osteoarthritis from October 2013 to October 2023 were collected. The outcome measures included visual analog scale, Western Ontario and McMaster Universities Osteoarthritis Index score, Timed Up and Go test, and 36-item short form health survey. Literature quality analysis was performed using the Cochrane Manual recommended tool for risk assessment of bias in randomized controlled trials. Two researchers independently completed the data collection, collation, extraction and analysis. RevMan 5.4 and Stata 18.0 software were used to analyze and plot the obtained data. RESULTS: A total of 29 articles with acceptable quality were included, involving 1 633 patients with knee osteoarthritis. The studies involved four types of exercise: aerobic training, strength training, flexibility/skill training, and mindfulness relaxation training. (1) The results of network meta-analysis showed that compared with routine care/health education, aerobic training could significantly improve pain symptoms (SMD=-3.26, 95%CI: -6.33 to -0.19, P < 0.05); strength training (SMD=-0.79, 95%CI: -1.34 to -0.23, P < 0.05) and mindfulness relaxation training (SMD=-0.79, 95%CI: -1.23 to -0.34, P < 0.05) could significantly improve the function of patients. Aerobic training (SMD=-1.37, 95%CI: -2.24 to -0.51, P < 0.05) and mindfulness relaxation training (SMD=-0.41, 95%CI: -0.80 to -0.02, P < 0.05) could significantly improve the functional mobility of patients. Mindfulness relaxation training (SMD=0.70, 95%CI: 0.21-1.18, P < 0.05) and strength training (SMD=0.42, 95%CI: 0.03-0.81, P < 0.05) could significantly improve the quality of life of patients. (2) The cumulative probability ranking results were as follows: pain: aerobic training (86.6%) > flexibility/skill training (60.1%) > strength training (56.8%) > mindfulness relaxation training (34.7%) > routine care/health education (11.7%); Knee function: strength training (73.7%) > mindfulness relaxation training (73.1%) > flexibility/skill training (56.1%) > aerobic training (39.9%) > usual care/health education (7.6%); Functional mobility: aerobic training (94.7%) > mindfulness relaxation training (65.5%) > strength training (45.1%) > flexibility/skill training (41.6%) > routine care/health education (3.2%); Quality of life: mindfulness relaxation training (91.3%) > strength training (68.0%) > flexibility/skill training (44.3%) > aerobic training (34.0%) > usual care/health education (12.3%). CONCLUSION: (1) Exercise therapy is effective in the treatment of knee osteoarthritis, among which aerobic training has the best effect on relieving pain and improving functional mobility. Strength training and mindfulness relaxation training has the best effect on improving patients’ function. Mindfulness relaxation training has the best effect on improving the quality of life of patients. (2) Limited by the quality and quantity of the included literature, more high-quality studies are needed to verify it. [ABSTRACT FROM AUTHOR]