<p>Despite recent attention to scientific productivity in the Sustainable Development Goals (SDGs), less is known about inequalities in achieving the goals, regarding research outputs and recognition, particularly in architecture. One of the key mechanisms reproducing these disparities is citation Matthew effect, reflecting the unequal distribution of citations in science, among different entities, including countries. Adopting a scientometric approach, SDG-related papers, published in architectural journals in 2021–2023, were extracted from SciVal. The contributing countries were clustered based on their Human Development Index and the Global Innovation Index, as proxies of development challenges and scientific and innovation capacity, respectively. The Matthew effect was calculated as the difference between a paper’s observed citations and the average citations of the related journal. Descriptive results revealed the existence of the Matthew effect, operating non-uniformly across all SDGs: While Industry, Innovation and Infrastructure (SDG9) and Decent Work and Economic Growth (SDG8) witnessed a positive Matthew effect, No Poverty (SDG1), and Clean Water and Sanitation (SDG6) experienced a negative value. Moreover, high-capacity, low-challenge countries dominated both productivity and citation gains. The GLMM confirmed the effect of publication type, SDGs, open access, number of authors, and affiliations on the magnitude of the effect. The significant SDG-country cluster interaction indicates the dependence of the citation advantage of specific SDGs on the national context. Overall, the global science system reproduces the self-reinforcing pattern of inequality embedded in the Matthew effect, even within research addressing global challenges. However, strategies such as open accessibility and fostering collaboration can mitigate these disparities.</p>