Purpose: In this study, we assessed whether there are independent associations between dietary total flavonoid intake and major flavonoid classes with retinal arteriolar and venular calibre. Methods: Blue Mountains Eye Study participants aged 49+ years who had complete data on diet and retinal vessel measures were analysed (n = 2821). Dietary intake was assessed using a semi-quantitative Food-Frequency Questionnaire (FFQ). Flavonoid content of foods in the FFQ was estimated using the US Department of Agriculture Flavonoid, Isoflavone and Proanthocyanidin databases. Fundus photographs were taken and retinal vascular calibre was measured using validated computer-assisted techniques. The associations of intake of dietary flavonoids with retinal vessel calibre were examined in linear regression models and general linear model. Results: The highest quartile of intake was compared with the lowest quartile using multivariable-adjustment models. Participants with the highest proanthocyanidin intake had narrower retinal venules (223.9 ± 0.62 versus 226.5 ± 0.63, respectively; Ptrend = 0.01); and the highest isoflavone intake was associated with wider retinal arterioles (188.1 ± 0.55 versus 186.3 ± 0.56, respectively; Ptrend = 0.01). The highest apple/pear consumption (a dietary source of catechin) was associated with narrower retinal venules (223.8 ± 0.57 versus 226.1 ± 0.52; Ptrend = 0.01) and wider retinal arterioles (187.9 ± 0.51 versus 186.2 ± 0.51; Ptrend = 0.02). Further, participants who were in the highest versus lowest quartile of chocolate consumption had ~ 2.1 μm narrower retinal venules (multivariable-adjusted P = 0.03). Conclusions: This study shows that higher intakes of specific flavonoid subclasses are associated with a favourable retinal microvascular profile. Greater consumption of flavonoid-rich apples/pears and chocolate was also associated with beneficial variations in retinal vascular calibre.