Although low-calorie diets (LCD) improve glucose regulation, it is unclear if interval exercise (INT) is additive. We examined the impact of an LCD versus LCD + INT training on ß-cell function in relation to glucose tolerance in obese adults. Twenty-six adults (Age: 46 ± 12 year; BMI 38 ± 6 kg/m2) were randomized to 2-week of LCD (~1200 kcal/day) or energy-matched LCD + INT (60 min/day alternating 3 min at 90 and 50% HRpeak). A 2 h 75 g oral glucose tolerance test (OGTT) was performed. Insulin secretion rates (ISR) were determined by deconvolution modeling to assess glucose-stimulated insulin secretion ([GSIS: ISR/glucose total area under the curve (tAUC)]) and ß-cell function (Disposition Index [DI: GSIS/IR]) relative to skeletal muscle (Matsuda Index), hepatic (HOMA-IR) and adipose (Adipose-IRfasting) insulin resistance (IR). LCD + INT, but not LCD alone, reduced glucose and total-phase ISR tAUC (Interactions: p = 0.04 and p = 0.05, respectively). Both interventions improved skeletal muscle IR by 16% (p = 0.04) and skeletal muscle and hepatic DI (Time: p < 0.05). Improved skeletal muscle DI was associated with lower glucose tAUC (r = −0.57, p < 0.01). Thus, LCD + INT improved glucose tolerance more than LCD in obese adults, and these findings relate to ß-cell function. These data support LCD + INT for preserving pancreatic function for type 2 diabetes prevention.