Objective To determine the median effective dose (ED50) and 95% effective dose (ED95) of remimazolam combined with alfentanil for painless gastroscopy. Methods A total of 60 patients who underwent gastroscopy were enrolled. According to their ages, they were divided into two groups (n=30): a young group (group Y, 18‒44 years old) and a middle‑aged group (group M, 45‒60 years old). All patients were intravenously infused with 5 μg/kg alfentanil, followed by a bolus injection of remimazolam 30 s later, with an initial dose of 0.1 mg/kg. When the modified Observer's Assessment of Alertness/Sedation Scale (mOAA/S) was less than 3 points, gastroscopy was performed. The current study was carried out using the up and down method, where the dose of remimazolam used by the next patient was determined based on the sedative effect of gastroscopy. If the patient "successfully" sedated, the remimazolam dose used by the next patient decreased by 0.03 mg/kg, otherwise increased by 0.03 mg/kg. The dose of remimazolam, gastroscopy time, recovery time, intraoperative awareness and adverse reactions were recorded. The ED50 and ED95 of remimazolam and their 95% confidence interval (CI) were calculated by the probit regression method. Results The ED50 of remimazolam was 0.13 (0.11, 0.16) mg/kg and the ED95 was 0.19 (0.16, 0.33) mg/kg in the group Y. The ED50 of remimazolam was 0.09 (0.07, 0.12) mg/kg and the ED95 was 0.18 (0.13, 0.57) mg/kg in the group M. The equivalent ratio of the group M versus the group Y was 0.68 (0.22, 0.98). There were no statistical differences in examination time, recovery time, the incidence of hypotension and respiratory depression between the two groups (P>0.05), and the dose of remimazolam in group Y was higher than that in group M (P<0.05). None of the patients had intraoperative awareness Conclusions When remimazolam is used in combination with 5 μg/kg alfentanil, the ED50 of remimazolam for painless gastroscopy is 0.13 (0.11, 0.16) mg/kg for young patients and 0.09 (0.07, 0.12) mg/kg for middle‑aged patients. The equivalent ratio of the two groups is 0.68 (0.22, 0.98).