Abstract Aims/Introduction To assess whether the sodium–glucose cotransporter 2 inhibitor, henagliflozin, improves cognitive impairment in patients with type 2 diabetes. Materials and Methods We carried out a prospective study on 290 patients with type 2 diabetes and cognitive impairment. Montreal Cognitive Assessment scores and plasma phosphorylated tau181 levels were used to assess cognition. The association between henagliflozin use and changes in cognition was examined using multivariable logistic regression analysis. Results Montreal Cognitive Assessment scores at enrollment and after 6 months were 21 (interquartile range [IQR]19–23) versus 22 (IQR 20–25; P 0.05) in the non‐sodium–glucose cotransporter 2 inhibitor group. Logistic regression analysis showed that henagliflozin treatment was associated with Montreal Cognitive Assessment score improvement independent of potential confounders (odds ratio [OR] 3.670, 95% confidence interval [CI] 2.224–6.056, P