Objective: Exercise programs at home are successful in treating gestational diabetes by controlling blood glucose. The aim is to compare the efficacy of the self-directed home exercise program, the standard care alone and the supervised home exercise program in pregnant women with gestational diabetes on blood glucose, quality of life and pregnancy outcomes. Methods: This randomized, parallel, single-blind study included 45 pregnant women who were 24–28 weeks of gestation. Participants were randomly divided into the supervised home exercise group (SHEG), home exercise group (HEG) and control group (CG). While the home exercises program was taught and controlled by a physiotherapist in SHEG, the home exercise brochure was given without any training by the gynecologist in HEG. Control group maintained their usual daily care. The home exercise intervention included low to moderate structured exercise performed three days per week for 8 weeks. Their glucose responses, quality of life and pregnancy outcomes were assessed pre- and post intervention. Results: Fasting glucose and 2 h postprandial glucose levels were improved statistically in SHEG and HEG groups after intervention (p < 0.05). Differences in SHEG were statistically higher than HEG (p < 0.017). When the HEG and CG were compared, there was no superiority between the two groups in all outcome measures except the physical health. Additionally, there were no statistically significant differences in values of cesarean birth and preterm birth between groups (p > 0.05). Conclusions: This study revealed that pregnant women should be under the supervision of physiotherapists while doing home exercises. Clinical Trial Registration The trial was approved by the registration of ClinicalTrials.gov and registration number: NCT05195333.