OBJECTIVE: Despite new recommendations for eight visits, four visits (ANC4+ ) remain low and inequitable in sub-Saharan Africa. Intra-urban estimates are often lacking to inform targeted interventions. This study applied small area estimation (SAE) to model the spatial variation of ANC4+ coverage in Addis Ababa city, Ethiopia. STUDY DESIGN: Cross-sectional. METHODS: We analyzed data from women who had given a live birth in the five years preceding the 2016 Ethiopia Demographic and Health Surveys (DHS) in Addis Ababa city (∼5million population). In addition to covariates from DHS, we included auxiliary data on travel time to the nearest health facility . A Bayesian hierarchical spatial model was used to model ANC4+ coverage in the 10 sub-cities of Addis Ababa city, adjusting for covariates. RESULT: Overall ANC4+ was 88.7 % (95 % CI: 84.2-92.1), with sub-city variation ranging from 80.6 % in Kolfe Keranio to 98.2 % in Nifas Silk Lafto. Eight of the ten sub-cities had ANC4+ coverage less than 90 %. Women less than 20 years (AOR = 0.30; 95 % CI: 0.99-1.16), with no education (AOR = 0.30; 95 % CI: 0.11-0.89), poorer wealth quintile (AOR = 0.40; 95 % CI: 0.19-1.04), and longer travel time (AOR = 0.90; 95 % CI: 0.82-0.99) were associated with lower ANC4+ utilization. Married women (AOR = 2.20; 95 % CI: 1.10-4.50) were associated with higher utilization of ANC4+. CONCLUSION: Despite Addis Ababa's relatively high overall ANC4+ coverage, significant differences at the sub-city level exist, with coverage still short of the 95 % global target. Targeted resource allocation and geographically informed policies are required to reduce these disparities.