The presence of glucocorticoids in healthy foods has recently become a topic of concern because of their side effects. In this study, we developed a method based on ultra-performance convergence chromatography–triple quadrupole mass spectrometry (UPC2–MS/MS) to detect 63 glucocorticoids in healthy foods. The analysis conditions were optimized, and the method was validated. We further compared the results of this method with those of the RPLC–MS/MS method. Glucocorticoids were separated on an Acquity Torus 2-picolylamine column (100 mm × 3.0 mm, 1.7 µm) and detected via MS/MS. CO2 and methanol (containing 0.1% formic acid) were used as mobile phases. The method demonstrated good linear relationships between 1 and 200 µg·L−1 (R2 ≥ 0.996). The limits of detection in different types of samples were 0.3–1.5 µg·kg−1 (S/N = 3). The average recoveries (n = 9) and RSDs in different types of samples were 76.6–118.2% and 1.1–13.1%, respectively. The matrix effect, calculated as the ratio between calibration curves built in matrix and pure solvent, was less than 0.21 for both a fish oil and a protein powder. This method exhibited better selectivity and resolution than RPLC–MS/MS method. Lastly, it could realize the baseline separation of 31 isomers of 13 groups, including four groups of eight epimers. This study provides new technical support for assessing the risk of exposure to glucocorticoids in healthy foods.