The Violence Risk Screening‐10 (V‐RISK‐10) is one of the few instruments available for evaluating violence risk among general psychiatric service users. This naturalistic prospective study involved 376 inpatients in a general psychiatric hospital in Beijing and intended to determine whether this brief instrument could be applied to a sample of Chinese consumers and whether its predictive properties could be retained. Risk assessment at admission was compared to the record of aggression and violence during the first month of hospitalization. During the research period, 108 of the 376 consumers caused 265 incidences of aggression. Receiver operating characteristics for the V‐RISK‐10 Chinese version yielded an area under the curve of 0.63. Its sensitivity/specificity was 0.80/0.38 and the corresponding positive/negative predictive value was 0.34/0.82. Intraclass correlation coefficient for the whole instrument was 0.89. Compared to the results of the original V‐RISK‐10, its predictive accuracy was lower. However, with some modification, the V‐RISK‐10 still shows promise as an instrument for use in daily practice in Chinese clinical settings.