Gestational diabetes mellitus results, in part, from a sub-optimal β-cell mass (BCM) during pregnancy. Artemisinins were reported to increase BCM in models of diabetes by α- to β-cell conversion leading to enhanced glucose tolerance. We used a mouse model of gestational glucose intolerance to compare the effects of an artemisinin (artesunate) on glycemia of pregnant mice with vehicle treatment (acetone) or no treatment. Animals were treated daily from gestational days (GD) 0.5 to 6.5. An intraperitoneal glucose tolerance test was performed prior to euthanasia at GD18.5 or post-partum. Glucose tolerance was significantly improved in both pregnant and non-pregnant mice with both artesunate and vehicle-alone treatment, suggesting the outcome was primarily due to the acetone vehicle. In non-pregnant, acetone-treated animals, improved glucose tolerance was associated with a higher BCM and a significant increase in bihormonal insulin and glucagon-containing pancreatic islet cells, suggesting α- to β-cell conversion. BCM did not differ with treatment during pregnancy or post-partum. However, placental weight was higher in acetone-treated animals and was associated with an upregulation of apelinergic genes. Acetone-treated animals had reduced weight gain during treatment despite comparable food consumption to non-treated mice, suggesting transient effects on nutrient uptake. The mean duodenal and ileum villus height was reduced following exposure to acetone. We conclude that acetone treatment may mimic transient fasting, resulting in a subsequent improvement in glucose tolerance during pregnancy.