There is no single hour when everyone should stop drinking caffeine. The caffeine half life—the time it takes your body to clear half the caffeine in your system—varies, and the dose matters too. A cautious starting point is to avoid large doses for at least six hours before bed; if sleep is a priority or you are sensitive to caffeine, leaving eight to twelve hours may be worth testing.
That is a practical range, not a guarantee. Research has found sleep effects from a large dose taken six hours before bedtime, and a newer trial found effects when a large dose was taken within twelve hours. A smaller dose did not have a significant effect in that newer trial, but the result cannot establish a safe cutoff for everyone.
What Caffeine Half Life Means For Your Bedtime
Caffeine blocks adenosine, a chemical involved in building sleep pressure while you are awake. Its effects do not stop as soon as you feel less alert. One review describes a typical half-life of about five hours, with estimates ranging from roughly 1.5 to 9.5 hours. In other words, caffeine can remain in your system for hours, but the rate it clears differs between people.
So a cup at 3 p.m. does not affect every person or every bedtime in the same way. Your total dose, individual metabolism and sensitivity all matter. A cutoff calculated from an average half-life can be a useful starting point, not a precise prediction of how you will sleep.
What Studies Say About Caffeine Before Bed
In a randomized crossover trial, Carissa Gardiner and colleagues tested 23 healthy men, with an average age of about 25, who usually consumed less than 300 mg of caffeine a day. Participants took 100 mg or 400 mg at different intervals—four, eight or twelve hours before bed—and researchers assessed sleep at home using monitoring and diaries. The 100 mg dose did not significantly change measured or reported sleep compared with placebo. The 400 mg dose delayed sleep and changed sleep patterns when taken within twelve hours; sleep fragmentation was greater when it was taken within eight hours.
Those results come from a small study of men, so they do not settle the question for all adults. A separate 2013 home-based study by Christopher Drake and colleagues found that 400 mg taken at bedtime or three or six hours beforehand disrupted sleep compared with placebo. Together, the trials suggest that a large dose can matter well before bedtime, while a lower dose may have a smaller effect. They do not prove that 100 mg will leave everyone’s sleep unaffected.
A 2025 review by Adem Kocak and colleagues pooled ten studies involving 128 athletes who consumed caffeine before late-afternoon or evening training or competition. The analysis found a small reduction in sleep efficiency, but this result was not robust in sensitivity checks. Athletes often reported worse sleep, while objective changes were inconsistent. The small number of studies and the mostly male participants limit how widely those findings can be applied.
Count More Than Your Coffee
Caffeine adds up across drinks, foods and supplements. As rough examples, an 8 fl oz (about 237 ml) coffee may contain around 80–100 mg, an 8 fl oz tea around 30–50 mg, and a 12 fl oz (about 355 ml) caffeinated cola around 30–40 mg. These amounts vary with the product, serving size and preparation; a “cup” is not a reliable measure across countries or cafés.
Chocolate also contributes caffeine, though the amount depends on the product and portion. For caffeine in pre-workout, check the label for caffeine per serving and how many servings you use. Some products may differ from the advertised dose, and caffeine from a supplement still counts alongside coffee, tea or cola. Labels and serving sizes differ by product and market, so use the specific package rather than assuming a standard amount.
Choose A Cutoff You Can Test
If you are deciding how long caffeine lasts for you, track your caffeine sources, amounts and timing alongside your usual bedtime and sleep experience for several days. Try moving your last serving earlier, especially if it is a large dose, and notice whether sleep changes. Feeling able to fall asleep does not necessarily mean caffeine had no effect: in Gardiner’s trial, perceived sleep quality and objective measures did not always align.
If you train late, weigh the alertness or performance benefit of caffeine against the possibility of poorer sleep; the athlete research does not offer a universal cutoff. Persistent sleep problems are worth discussing with a healthcare professional, particularly if changing caffeine timing does not help.
