Objective To compare the short-term neurological prognosis and rebleeding risk among patients with supratentorial hypertensive intracerebral hemorrhage （HICH） who were treated with neuroendoscopic hematoma evacuation （neuroendoscopy group）， soft-channel puncture drainage with urokinase infusion （puncture group）， and craniotomy for hematoma evacuation （craniotomy group）. Methods A total of 384 patients with supratentorial HICH who were admitted to the Department of Neurosurgery， Huai'an First People's Hospital Affiliated to Nanjing Medical University between January 2021 and December 2025 were retrospectively enrolled. These patients were assigned to the craniotomy group （n = 121）， neuroendoscopy group （n = 107）， and puncture group （n = 156） based on the surgical modality. The inclusion criteria were a hematoma volume ranging from 30 to 80 mL and an admission Glasgow coma scale （GCS） score between 5 and 12. Three-category propensity score overlap weighting （PS-OW） was employed to balance the baseline characteristics across the groups. The primary outcome was a poor short-term prognosis， which was defined as a modified Rankin scale （mRS） score of 4 or higher at the time of discharge. Secondary outcomes included the rebleeding rate， the GCS score at discharge， the length of hospital stay， and the total hospitalization cost. Results After weighting， the baseline covariates were well-balanced among the three groups （all absolute standardized mean differences < 0.1）. The weighted event rate of poor prognosis in the craniotomy group was 22.8%， which was higher than the 11.0% in the neuroendoscopy group and the 11.5% in the puncture group. When compared with the neuroendoscopy group， the craniotomy group exhibited a significantly higher risk of poor prognosis （odds ratio ［OR］=2.389， 95%CI： 1.029 - 5.545， P = 0.043）. A similar trend was noted when comparing the craniotomy group with the puncture group （OR = 2.280， 95%CI： 0.982 - 5.290， P = 0.055）. No statistically significant difference in the rebleeding rate was detected among the three groups. The craniotomy group had the longest hospital stay and the highest total cost， whereas the puncture group had the lowest. The neuroendoscopy group presented a slightly better GCS score at discharge than the craniotomy group. Conclusions In the overlapping population eligible for all three surgical modalities， craniotomy was associated with a poorer short-term neurological prognosis and a greater medical burden compared with minimally invasive approaches. Meanwhile， no clinically significant difference in rebleeding risk was observed among the three procedures.