Divergent thinking (DT) tests are sometimes used to select students for gifted programs. Studies on these tests, mostly conducted on non-gifted students, suggest that performance is influenced by the type of instruction given (standard vs. hybrid "be fluent AND original") and time-on-task. The current study aimed to examine the effect of instructions and time-on-task on divergent thinking performance in gifted and non-gifted students in a 2 [gifted versus non-gifted] × 2 [standard versus hybrid instructions] design. The results showed that gifted students outperformed non-gifted students in fluency, while no significant difference was found between the two groups in originality. Creativity instructions improved both originality and fluency scores in verbal but not figural tests. As for time-on-task, gifted students took more time when completing DT tests as well as when they were given explicit instructions to "be creative." Implications for gifted identification are discussed.