Aims/Introduction The predictive value of admission hyperglycemia in the long‐term prognosis of acute myocardial infarction patients is still controversial. We aimed to investigate this value based on the diabetes status. Materials and Methods We carried out a multicenter, retrospective study of 1,288 acute myocardial infarction patients enrolled in 11 hospitals between March 2014 and June 2019 in Chengdu, China. The patients were classified into those with diabetes and those without diabetes, each was further divided into: hyperglycemia and non‐hyperglycemia subgroups, according to the optimal cut‐off value of the blood glucose to predict all‐cause mortality during follow up. The end‐points were all‐cause death and major adverse cardiovascular and cerebrovascular events, including all‐cause death, non‐fatal myocardial infarction, vessel revascularization and non‐fatal stroke. Results In the follow‐up period of 15 months, we observed 210 (16.3%), 6 (0.5%), 57 (4.4%) and 34 (2.6%) cases of death, non‐fatal myocardial infarction, revascularization and non‐fatal stroke, respectively. The optimal cut‐off values of admission blood glucose for patients with diabetes and patients without diabetes to predict all‐cause mortality during follow up were 14.80 and 6.77 mmol/L, respectively. We divided patients with diabetes (n = 331) into hyperglycemia (n = 92) and non‐hyperglycemia (n = 239), and patients without diabetes (n = 897) into hyperglycemia (n = 425) and non‐hyperglycemia (n = 472). The cumulative rates of all‐cause death and major adverse cardiovascular and cerebrovascular events among the patients in each hyperglycemia group was higher than that in the corresponding non‐hyperglycemia group (P < 0.001). In patients without diabetes, admission hyperglycemia was an independent predictor of all‐cause mortality and major adverse cardiovascular and cerebrovascular events. Conclusion Admission hyperglycemia was an independent predictor for long‐term prognosis in acute myocardial infarction patients without diabetes. The prognostic value of admission hyperglycemia on long‐term prognosis remains elusive, especially in acute myocardial infarction patients with diabetes. Regarding acute myocardial infarction patients, there has been no accurate threshold of admission hyperglycemia to predict mortality. Therefore, in our study, we used different cut‐off values in patients with and without diabetes to discuss their predictive value in the long‐term prognosis of acute myocardial infarction patients.