It is reported that miR-26a-5p could regulate neuronal development, but its underlying mechanisms in Alzheimer's disease (AD) progression is unclear. APP (swe)/PS1 (ΔE9) transgenic mice served as AD mice. Morris water maze test was used to measure the spatial learning and memory ability of mice. The expressions of miR-26a-5p, DYRK1A, phosphorylated-Tau, Aβ40, and Aβ42 were detected. The relationship between miR- 26a-5p and DYRK1A was explored using dual luciferase reporter assay. The effects of miR-26a- 5p on AD mice was determined. AD mice walked a lot of wrong ways to find the platform area and the latency time to reach the platform was longer. There was low expression of MiR-26a-5p in AD mice. Overexpression of miR-26a-5p inhibited Tau phosphorylation and Aβ accumulation. MiR-26a-5p negatively regulated DYRK1A via targeting its 3'UTR. In vivo, increased miR-26a-5p down-regulated Aβ40, Aβ42, p-APP and p-Tau levels in AD mice through decreasing DYRK1A. Meanwhile, the swimming path and the latency time, to reach the platform, was shorten after enhancing miR-26a-5p expression. Overexpression of miR-26a-5p could repress Tau phosphorylation and Aβ accumulation via down-regulating DYRK1A level in AD mice.