Normative models of brain morphometry quantify individual deviations from typicalanatomical patterns and hold promise for enhancing clinical decision-making. However,their clinical utility depends critically on demonstrating generalizability across diverseethnoracial populations. We previously developed sex-specific, race-neutral normativemodels for cortical thickness, surface area, and subcortical volumes using brain scansfrom a large international sample of healthy individuals, as part of the CentileBrainProject, a global initiative to provide open-access, neuroimaging reference models.The primary aim of the present study was to empirically evaluate the generalizabilityand accuracy of these pretrained models across multiple ethnoracial groups. To thisend, we tested model performance in independent samples of healthy individuals fromAfrica, Asia, Europe, and the Americas, with ethnoracial classification defined eitherby self-identification or genetic ancestry (N = 4,862). We further compared perfor-mance against normative models developed exclusively from a single-population Chinesecohort. Across all groups, as well as in the pooled sample, the pretrained CentileBrainmodels demonstrated consistently high accuracy, with relative mean absolute error valuesbelow 10% for subcortical volume and surface area and below 5% for cortical thickness.Model performance was highly concordant across self-identified and ancestry-definedgroups. In a separate analysis, the CentileBrain models performed comparably to apopulation-specific model when applied to an independent ancestry-matched sample.These findings provide empirical support for the generalizability of race-neutral norma-tive models developed on large and diverse samples and underscore their potential utilityfor individualized neuroimaging assessment across ethnoracially diverse populations.