Limited research has examined the effect of meal composition on sleep. Based on previous research, we hypothesized that a low glycemic index (LGI) drink containing 50 g isomaltulose (Palatinose, GI = 32) would result in more N3 sleep, less rapid eye movement (REM) sleep, and better memory consolidation than a high glycemic index (HGI) drink containing 50 g glucose (GI = 100). Healthy males (n = 20) attended the laboratory on three occasions at least a week apart (one acclimatization night and two test nights). Using a repeated measures, randomized, double-blind design, participants consumed a standardized evening meal followed four hours later by a HGI or LGI drink. Sleep architecture and continuity were assessed using polysomnography. Procedural and episodic memory were assessed pre- and post-sleep using a finger tapping task and story recall task, respectively. There was no main effect of drink. However, there was an interaction between drink and drink order. N3 sleep percentage was significantly longer (28.71% vs 23.99%, respectively, p = .032) and overnight retention of neutral story content was significantly better (0.63 words vs -10.13 words, respectively, p = .002) after the LGI drink than HGI drink, but only when the LGI drink was consumed on the second test night and HGI drink on the first test night. No changes in REM sleep were observed. Findings suggest that the nature of carbohydrate consumed before bed may influence sleep quality and quantity and neutral episodic memory consolidation. Pre-registered with ClinicalTrials.gov (NCT05591573).