Aims/hypothesis The objective of this randomised crossover study was to determine, in adults with type 1 diabetes, whether the glycaemic effects of high-intensity interval exercise (HIIE) vs moderate-intensity continuous exercise (CONT) differ between the postprandial (PP, higher insulin-on-board) and postabsorptive (PA) states. Methods Twenty physically active adults (inclusion criteria: type 1 diabetes for textgreater1 year, 18–55 years old; nine female; mean ± SD age 33.4 ± 9.4 years) completed four iso-mechanical 25 min cycling sessions (HIIE-PA and HIIE-PP including 1’-1’ bouts at 100% maximal aerobic power [MAP]; CONT-PA and CONT-PP at 50% MAP) in randomised order, at the EURASPORT facility of Lille University. Capillary glucose (primary outcome), lactate, cardiorespiratory variables, perceived exertion and enjoyment were assessed during exercise. Interstitial glucose, dietary intake and insulin administration were recorded over 24 h post-exercise. Mixed models assessed effects of exercise modality, prandial state, and their interaction on glycaemic excursions. Results Despite similar mechanical load, HIIE induced a greater physiological response than CONT (higher oxygen consumption, heart rate, blood lactate concentration and perceived exertion; ptextless0.05) without reducing exercise enjoyment. During exercise, capillary glucose declined less in the PA state than in the PP state (p=0.003) and less during HIIE than during CONT (p=0.038), with the protective effect of HIIE being more pronounced in the PA state (p=0.001) (HIIE-PA, −0.72 ± 1.71 mmol/l; CONT-PA, −1.73 ± 1.48 mmol/l; HIIE-PP, −2.33 ± 1.61 mmol/l; CONT-PP, −2.61 ± 1.27 mmol/l). Over the 24 h post-exercise period, participants spent less time at textless3.0 mmol/l following HIIE than following CONT (after adjustment for carbohydrate intake and insulin doses) despite a lower dietary glycaemic load.