Rationale: The 5-choice serial reaction time task (5-CSRTT) is a widely used operant task for measuring attention and motor impulsivity in rodents. Training animals in this task requires an extensive period of daily operant sessions. Recently, a self-paced, automated version of this task has been developed for mice, which substantially reduces training time. Whether a similar approach is effective for rats is currently unknown. Objective: Here, we tested whether attention and impulsivity can be assessed in rats with a self-paced version of the 5-CSRTT. Methods: Operant boxes were connected to home-cages with tunnels. Two groups of rats self-paced their training by means of an automated script. The first group of animals was allowed unlimited access (UA) to start trials in the task; for the second group, trial availability was restricted to the first 2.5 h of the dark cycle (TR). Task parameter manipulations, such as variable inter-trial intervals and stimulus durations as well as pharmacological challenges with scopolamine, were tested to validate the task. Results: Self-paced training took less than 1 week. Animals in the UA group showed higher levels of omissions compared with the TR group. In both protocols, variable inter-trial intervals increased impulsivity, and variable stimulus durations decreased attentional performance. Scopolamine affected cognitive performance in the TR group only. Conclusions: Home-cage-based training of the 5-CSRTT in rats, especially the TR protocol, presents a valid and fast alternative for measuring attention and impulsivity.