Background: Few studies have investigated the association between overall diet and cardiorespiratory fitness (CRF). Objective: We aimed to investigate associations of food groups, a diet-quality score, and dietary patterns with CRF in black and white adults. Design: We included 2632 participants aged 38-50 y who attended the year 20 exam of the Coronary Artery Risk Development in Young Adults (CARDIA) study and Coronary Artery Risk Development in Young Adults Fitness Study (CFS) conducted in 2005-2006. Diet was assessed by using a validated diet history. A dietary score and 2 types of patterns were included as follows: the a priori diet-quality score and meat and fruit-vegetable dietary patterns derived from principal components analysis. CRF was assessed by using a graded exercise treadmill test. Linear regression models regressed the treadmill duration on food groups and dietary scores and patterns overall and in race-sex subgroups. Results: Grains (whole and refined), processed meats, and beverages (coffee, meal-replacement drinks, beer, and wine) were positively associated with the treadmill duration overall; whole fruit (not juices), organ meats, fried meats and fish, and soy and nondairy products were negatively associated. The a priori diet-quality score was positively associated with the duration overall and in all racesex subgroups (P <0.05) except black men. The meat pattern was negatively associated with the duration in white men and white women only. The fruit-vegetable pattern was positively associated with duration in white women only. Conclusions: Overall, the a priori diet-quality score was positively associated with CRF in this cohort of black and white adults, whereas the meat dietary pattern was negatively associated only in whites. The CARDIA study and CFS were registered at clinicaltrials.gov as NCT00005130 and NCT00106457, respectively.