Stunting is a major public health problem affecting more than one-third of under five year's old children in Ethiopia. It has short and long (irreversible) consequences, including stunted growth, never reaching physical and cognitive potential, struggles in school, and increased morbidity and mortality due to infections. Though stunting is the leading cause of child mortality in Ethiopia, evidence is scarce on the prevalence and predictors of stunting among under-five years old children in Ethiopia. Therefore, this study aimed to estimate the prevalence and predictors of stunting severity among under-5 children in Ethiopia. This study was based on 2019 Mini-Ethiopian Demographic and Health Survey (EDHS) data. A weighted total sample of 4972 under-five years old children was included in the study. Height measurement was collected for each child. Anthropometric indicator, height-for-age was determined for children using World Health Organization growth standards (Z-scores for Height-for-Age (HAZ)) to asses stunting level. Given the ordinal nature of stunting and the hierarchical nature of EDHS data, a multilevel ordinal logistic regression model was applied. Brant test was used to check the proportional odds assumption, which was satisfied (P-value ≥0.05). Moreover, deviance was used for model comparison. For the multivariable analysis, variables with a p-value ≤0.2 in the bivariable analysis were considered. The Adjusted Odds Ratio (AOR) with 95% Confidence Interval (CI) was reported as associated factor to the severity levels of stunting in the multivariable multilevel proportional odds model. The overall prevalence of stunting among under-5 children in Ethiopia was 35.7% [95% CI: 34.4%, 37.1%]. Of these, 12.1% were severely stunted, and 24.9% were moderately stunted. Being male [AOR = 0.83, 95% CI: 0.74, 0.93], children aged 6-23 months [AOR = 2.38, 95% CI: 1.84, 3.07], ≥ 24 months [AOR = 4.15, 95% CI: 3.26, 5.28], children whose maternal age 15-24 years [AOR = 0.73, 95% CI: 0.58, 0.92], children from the poorest, poorer, middle, and richer household wealth were [AOR = 1.84, 95% CI: 1.32, 2.57], [AOR = 1.66, 95% CI: 1.20, 2.31], [AOR = 1.78, 95% CI: 1.29, 2.44], and [AOR = 1.62, 95% CI: 1.20, 2.17], children whose maternal educational status of no formal education and primary education had [AOR = 1.90, 95% CI: 1.28, 2.82], [AOR = 1.78, 95% CI: 1.22, 2.60], Tigray [AOR = 2.95, 95% CI: 1.78, 4.86], Afar [AOR = 1.85, 95% CI: 1.11, 3.10], Amhara [AOR = 1.90, 95% CI: 1.14, 3.14] and Harari [AOR = 1.97, 95% CI: 1.20, 3.25]regions, low community maternal education [AOR = 0.76, 95% CI: 0.62, 0.92] were significantly associated with stunting severity levelling. Stunting among children under five years of old in Ethiopia remains a major public health issue. Improving access to maternal education is related to appropriate child feeding practices and health, particularly in younger and uneducated mothers. Strengthening the family's wealth status is also recommended to reduce stunting. In addition, it is better to support strategies of preconception care for mothers during pregnancy to reduce stunting in the long term.