Objective To explore the optimal dose of remimazolam combined with remifentanil in painless gastroscopy. Methods It was a randomized, single⁃blind, controlled study. A total of 699 patients who underwent painless gastroscopy, According to the random number table method, they were divided into seven groups. Groups A (n=100), B (n=100), and C (n=100) were treated with 0.2 mg/kg, 0.3 mg/kg or 0.4 mg/kg of remimazolam combined with 0.25 μg/kg of remifentanil, respectively. Groups D (n=100), E(n=100) and F (n=100) were treated with 0.2 mg/kg, 0.3 mg/kg and 0.4 mg/kg of remifentanil combined with 0.50 μg/kg of remifentanil. Group R (n=99) was administered with 2 mg/kg of propofol. Their sedation success rate, circulatory and respiratory system indicators, incidence of adverse reactions, recovery time and quality, andsatisfaction of patients and operating doctors were compared. We recorded the incidence of hiccups between 3 groups of remifentanil 0.25 μg/kg, 3 groups of remifentanil 0.50 μg/kg, 2 groups of remimazolam 0.2 mg/kg, 2 groups of remimazolam 0.3 mg/kg, 2 groups of remimazolam 0.4 mg/kg and analyzed the correlation between two drug doses and hiccups. Results The sedation success rate of group A, B, C, D, E, F was higher than that of group R (P<0.05). The success rate of sedation increasing shows dose dependence of remimazolam when the dose of remifentanil is constant, or dose dependen cy of remifentanil when the dose of remimazolam is constant. The heart rate and mean arterial pressure (MAP) fluctuations of patients during gastroscopy were reduced with the increased sedation rate. Also, with the increased sedation rate, the volatility and the incidence of coughing and body movement decreased while the satisfaction scores of patients and operators increased (P<0.05). However, with the increased sedation extent, the incidence of respiratory depression and hypotension were also increased while the amplitude of heart rate decreasing and MAP increased. The patient's time to wake up and leave the hospital were prolonged (P<0.05). There was a significant weak positive correlation between the dose of remimazolam and the occurrence of hiccups (0.25 μg/kg, r=0.180, P<0.05;0.50 μg/kg, r=0.272, P<0.05), while a significant negative correlation between the dose of remifentanil and the occurrence of hiccups (0.2 mg/kg, r=−0.400, P<0.05; 0.30 mg/kg, r=−0.386, P<0.05; 0.40 mg/kg, r=−0.303, P<0.05). Conclusions The combined use of remimazolam and remifentanil is safe during painless examination, and 0.3 mg/kg of remimazolam combined with 0.25 μg/kg of remifentanil is an appropriate dose combination.