Medical nutrition therapy is an integral part of gestational diabetes mellitus (GDM) management; however, the prescription of optimal energy intake is often a difficult task due to the limited available evidence. The present pilot, feasibility, parallel, open-label and non-randomized study aimed to evaluate the effect of a very low energy diet (VLED, 1600 kcal/day), or a low energy diet (LED, 1800 kcal/day), with or without personalized exercise sessions, among women with GDM in singleton pregnancies. A total of 43 women were allocated to one of four interventions at GDM diagnosis: (1) VLED (n = 15), (2) VLED + exercise (n = 4), (3) LED (n = 16) or (4) LED + exercise (n = 8). Primary outcomes were gestational weight gain (GWG), infant birth weight, complications at delivery and a composite outcomes score. Secondary outcomes included type of delivery, prematurity, small- for-gestational-age (SGA) or large-for-gestational-age (LGA) infants, macrosomia, Apgar score, insulin use, depression, respiratory quotient (RQ), resting metabolic rate (RMR) and middle-upper arm circumference (MUAC). GWG differed between intervention groups (LED median: 12.0 kg; VLED: 5.9 kg). No differences were noted in the type of delivery, infant birth weight, composite score, prevalence of prematurity, depression, RQ, Apgar score, MUAC, or insulin use among the four groups. Regarding components of the composite score, most infants (88.4%) were appropriate-for-gestational age (AGA) and born at a gestational age of 37–42 weeks (95.3%). With respect to the mothers, 9.3% experienced complications at delivery, with the majority being allocated at the VLED + exercise arm (p < 0.03). The composite score was low (range 0–2.5) for all mother-infant pairs, indicating a “risk-free” pregnancy outcome. The results indicate that adherence to a LED or VLED induces similar maternal, infant and obstetrics outcomes.