Existing mouse models of Roux-en-Y gastric bypass (RYGB) surgery are not comparable to human RYGB in gastric pouch volume for a large or absent gastric volume. The aim of this study was to develop and characterize a mouse RYGB model that closely replicates gastric pouch size of human RYGB surgery of about 5% of total gastric volume. We established this model in diet-induced obese (DIO) mice of C57BL/6J. This surgery resulted in a sustained 30% weight loss, entirely accounted for by decreased fat mass but not lean mass, compared to sham-operated mice on the high fat diet. Compared to sham-operated mice, energy expenditure corrected for total body weight was significantly increased by about 25%, and substrate utilization was shifted toward higher carbohydrate utilization at 8 weeks after RYGB when body weight had stabilized at the lower level. The energy expenditure persisted and carbohydrate utilization was even more pronounced when the mice were fed chow diet. Although significantly increased during daytime, overall locomotor activity was not significantly different. In response to cold exposure, RYGB mice exhibited an improved capacity to maintain the body temperature. In insulin tolerance test, exogenous insulin-induced suppression of plasma glucose levels was significantly greater in RYGB mice at 4 weeks after surgery. Paradoxically, food intake measured at 5 weeks after surgery was significantly increased, possibly in compensation for increased fecal energy loss and energy expenditure. In conclusion, this new model is a viable alternative to existing murine RYGB models and the model matches human RYGB surgery in anatomy. This model will be useful for studying molecular mechanisms involved in the beneficial effects of RYGB on body weight and glucose homeostasis.