Levonorgestrel intrauterine system (LNG-IUS) has been widely used in patients with endometrial carcinoma (EC), endometrial hyperplasia without atypical (EH), and atypical endometrial hyperplasia (AEH). The purpose of our Network meta-analysis (NMA) is to evaluate the efficacy of the treatments based on the LNG-IUS in patients with EC and EH with or without atypical. We examined PubMed, EMBASE, Web of Science and the Cochrane Library up to 22 April 2024 to determine studies reporting treatment outcomes in EC and EH patients receiving LNG-IUS therapy, LNG-IUS + metformin (MET), oral progestins (OP), etc. We used EndNote 9 to select studies, Jadad scale and NOS scale to assess quality, stata(16.0) and R (4.3.1) to analysis the data. Overall, 28 studies involving 3752 patients were included in our NMA. As for EH patients, LNG-IUS (RR 1.21; 95% CrI [1.11, 1.34]) and LNG-IUS + MET (RR 323.57; 95% CrI [1.61, 214,223,188.1])] significantly increased CR rate in comparison with OP. Based on SUCRA, LNG-IUS + OP was the best treatment to improve CR(SUCRA = 67.2%) in patients with EC, whereas LNG-IUS + MET was superior in increasing CR (SUCRA = 99.8%) than any other treatments for EH patients. Besides, the ranking based on SUCRA illustrated that LNG-IUS alone was the best choice to raise CR rates (SUCRA = 76.7%) for AEH patients. In head-to-head meta-analysis, OP has a higher progression rate (RR 4, 95% CI 1.89-8.46, p = 0.062; I2 = 71.3%), a higher nausea rate (RR 1.93, 95% CI 1.24-3.01, p = 0.187; I2 = 40.4%) than LNG-IUS in patients with EH. In contrast, LNG-IUS had a irregular vaginal bleeding rates (RR 0.76, 95% CI 0.64-0.90, p = 0.034; I2 = 77.7%) than OP in EH patients. In addition, as for AEH patients, OP has a higher persistence rate (RR 4.31, 95% CI 1.43-13.00, p = 0.93; I2 = 0.0%) than LNG-IUS. According to the NMA, LNG-IUS related studies are feasible for conservative therapy in patients with EC and EH with or without atypical. Therefore, concerning the curative effect, we recommend LNG-IUS-based treatments as the best conservative therapy for EC and EH patients. However, future studies require large sample sizes and more outcomes to further evaluate the differences of treatment selections based on LNG-IUS.