It is commonly asserted that snacking between meals can lead to weight gain. However, there is little evidence for a positive relationship between snacking and obesity in humans. Our aim was to investigate compensatory behaviour in male and female rats given a palatable, rewarding food snack. We hypothesised that homeostatic systems controlling energy balance would protect from weight gain, initiating compensatory responses such as a reduction in intake of other food sources. Adult male or female Sprague Dawley rats were housed individually in standard conditions with ad lib access to bland food and water. Rats received restricted access to sweetened condensed milk (15 min/day for 10 days). In response to daily reward consumption (~24% of total voluntary daily kcal intake) males compensated accurately by reducing voluntary bland food intake (19±4%, n=7), however, female rats only partially compensated by reducing voluntary food intake (13±4%, n=8). During the reward access period there was no correlation between body weight and total daily kcal intake in males (R2=0.0042, p=0.64), but a positive correlation was seen in females (R2=0.12, p=0.0003). To test whether there is a limit to compensation in males, rats were presented with three daily food rewards (~56% of voluntary daily kcal intake). Male rats failed to compensate fully, reducing their voluntary food intake by only 35±7% (n=7) leading in turn to an increase in total kcal intake. Despite this increased intake, male rats did not show an increase in body weight and there was no positive correlation between body weight gain and total daily kcal intake (R2=0.0018, p=0.79). To test whether learning plays a role in compensation, male rats were presented with irregular food rewards (0-4 rewards/day, presented between 09.00 and 18.00 for 14 days). In this unpredictable access paradigm, rats were still able to compensate fully for reward kcal and maintained a similar body weight to rats fed the same total number of rewards on a regular basis. Our data show that male rats compensate well for small food rewards but females do not and quickly gain weight. The basis for this sex difference is not known. Male rats do not need to learn in order to compensate. Instead, homeostatic mechanisms seem to track total daily kcal consumption despite a reward-driven bias towards consuming high-energy, poor nutrition foods. However, males do not compensate completely for small food rewards. It is possible that homeostatic requirements for other macronutrients drive bland food intake despite the reward-driven overconsumption of kcal.