We explored how the rated beauty of an abstract painting is influenced by the relative beauty of a context painting. To this end, average-beauty abstract target paintings were paired with either a low-beauty or a high-beauty context painting. This context-beauty manipulation was crossed with 3 other between-groups factors common in studies of contrast/assimilation effects: (a) context paintings were either of a similar (abstract) or different (representational) style, (b) context-target pairs were presented either sequentially or simultaneously, and (c) participants rated either 1 or 5 context-target pairs. Abstract paintings were deemed more beautiful when paired with the low-beauty (vs. high-beauty) paintings, and this contrast effect was not moderated by our other manipulations—even though each yielded a main effect. This ubiquitous contrast pattern challenges current selective accessibility and range-frequency models of context effects on aesthetic judgments. To accommodate our findings, these models will need to better specify how participants perceive and assess the similarity of pairs of paintings.