Objective: To analyze the risk factors of intracranial hemorrhage (IH) in patients with acute ischemic stroke (AIS) treated with intra-arterial thrombectomy (IAT) in extended time window, and to establish the predictive model for guiding clinical decision. Materials and methods: From January 2018 and June 2020, clinical data of 146 AIS patients treated with IAT in extended time window were retrospectively analyzed. Patients were divided into hemorrhage and no-hemorrhage group. Demographic, clinical and imaging data were compared between two groups. Logistic regression (LR) analysis was applied to clarify the risk factor of IH after IAT, and to establish the predictive model. Receiver operator characteristic curve (ROC) analysis was used to evaluate the performance of the model for predicting IH after IAT. Results: Intracranial hemorrhage occurred in 48 (32.9%) patients. Compared with no-hemorrhage group, hemorrhage group showed higher baseline National Institute of Health Stroke Scale (NIHSS) (p<0.001), higher numbers of stent thrombectomy (p=0.049) and lower baseline alberta stroke program early computed tomography score (ASPECTS) (p<0.001). LR analysis indicated that, high NIHSS (p=0.001) and low ASPECTS (p<0.001) were risk factors of IH. The predictive model was as follows: hemorrhage risk value = -0.535 + 0.130 baseline NIHSS - 0.597 ASPECTS. The predictive model showed an area under the ROC curve, sensitivity and specificity of 0.875, 0.854 and 0.837, respectively. Conclusions: In the AIS patients in extended time window, the patients with high baseline NIHSS and low ASPECTS were prone to IH after IAT. The predictive model can provide reference for prevention of IH and clinical observation.