Certain phthalates and bisphenol A (BPA) have been associated with insulin resistance and type 2 diabetes in non-pregnant adults, but studies of gestational diabetes mellitus (GDM) have reported conflicting results for phthalates and no associations with BPA. Our aim was to investigate the relationship between maternal serum levels of phthalate metabolites and phenols at 10-17 weeks of gestation and glucose homeostasis at 28 weeks of gestation. 232 women aged ≥16 years without type 1 or 2 diabetes with singleton male pregnancies were recruited from a single UK maternity centre between 2001 and 2009 as part of a prospective observational study (Cambridge Baby Growth Study). Serum levels of 16 phthalate metabolites and 9 phenols (including BPA) were measured using liquid chromatography/tandem mass spectrometry. Oral glucose tolerance tests were performed at 28 weeks. 47/232 (20.3%) women had GDM. First-trimester triclosan (TCS) was inversely associated with incident GDM (adjusted odds ratio per log increase in concentration 0.54, 95% confidence interval 0.34-0.86, p = 0.010). Amongst women without GDM, first-trimester mono-(2-ethylhexyl) phthalate and mono(carboxyisooctyl) phthalate levels were positively associated with 120-min plasma glucose (adjusted β 0.268 and 0.183, p = 0.0002 and 0.010, respectively) in mid-pregnancy. No other monotonic associations were detected between phthalate or phenol levels and fasting or stimulated plasma glucose, β-cell function, insulin resistance, or 60-min disposition index. Our results support a glycaemia-raising effect of phthalates during pregnancy, consistent with findings in non-pregnant populations and suggest a possible protective effect of exposure to TCS against GDM.