None of the three is a sleeping pill, and none has been tested against the other two. Magnesium has the largest body of evidence, with a small effect on self-reported insomnia. Glycine has a few encouraging but tiny Japanese trials. GABA capsules have the weakest evidence, and we do not even know whether it reaches the brain.
The trio has become a fixture in “sleep stacks” with an appealing promise: simple nutrients that “regulate” your brain rather than knock it out. It is a good sales pitch. It remains a hypothesis until trials actually measure it.
Before choosing a capsule: what is your problem?
“Sleeping badly” covers very different situations, and a supplement cannot address them all:
- you take more than 30 minutes to fall asleep
- you wake during the night and struggle to get back to sleep
- you sleep long enough but wake up tired
- you get too little sleep because you go to bed too late
In the last case, no compound will make up for too short a sleep window. In the third, sleep apnea · FR or accumulated sleep debt need to be ruled out before you buy anything. The sleep analyzer helps you define the problem.
Magnesium: the largest evidence base, the most modest effect
Magnesium helps regulate nerve excitability. That is a real mechanism, but it does not tell us that a supplement improves sleep in someone whose intake is already adequate.
The trials:
- An Iranian trial from 2012 in 46 older adults with insomnia gave 500 mg of magnesium daily for eight weeks. Insomnia scores, time to fall asleep and cortisol decreased in the magnesium group, with differences mainly in self-reported outcomes.
- A 2021 systematic review found only three trials, totaling 151 older adults. Participants fell asleep 17 minutes faster on average, but the difference was not statistically significant, and the evidence was rated low to very low quality.
- A more recent, better-designed 2025 trial of bisglycinate found a 1.6-point difference versus placebo on an insomnia scale ranging from 0 to 28. Details are in the magnesium bisglycinate guide.
Translation: a small signal, mainly in older adults or people who sleep poorly, not a transformed night’s sleep. Magnesium has the advantage of being inexpensive and well understood, and the drawback of causing digestive problems at high doses.
Glycine: promising, tiny Japanese trials
Glycine is the body’s smallest amino acid. In rats, it acts on the brain’s central clock and promotes a drop in body temperature, which normally accompanies falling asleep. This mechanism was described in a 2015 study, in animals.
In humans, the evidence rests on a few small Japanese trials using 3 g before bed:
- a 2007 crossover trial reported better self-reported sleep quality, with consistent changes in sleep recordings
- a 2012 trial in volunteers deprived of a quarter of their sleep for three nights found less sleepiness and fatigue the next day, and better alertness on a reaction-time test
These studies are small, Japanese, and partly linked to a glycine manufacturer. The signal is consistent, not established. Glycine is, however, well tolerated at these doses in healthy adults; precautions are detailed in the guide to glycine side effects.
GABA: the biggest seller, the least proven
GABA is the brain’s main inhibitory neurotransmitter. Benzodiazepine-type sleeping pills act on its receptors. That is precisely what makes it marketable: “your brain’s natural brake, in a capsule.”
The problem has two parts. First, we do not know whether swallowed GABA crosses the barrier protecting the brain. Second, trials are scarce. A 2020 review that included only placebo-controlled trials concluded that evidence was limited for stress and very limited for sleep.
The most cited trial, published in 2018, gave 300 mg of rice-germ-derived GABA to 30 people reporting insomnia, versus 10 on placebo, for four weeks. Measured time to fall asleep fell from 13 to 6 minutes in the GABA group. Two caveats: the comparison was made within each group rather than between groups, and a baseline time to fall asleep of 13 minutes is hardly sleep-onset insomnia.
It is a textbook example of Mitochondria Marketing for sleep: a real mechanism (GABA slows neurons down) turned into a product promise (the capsule slows yours down), without evidence connecting the two.
Comparison: which to try for your situation
| Magnesium | Glycine | Oral GABA | |
|---|---|---|---|
| Human evidence | Several small trials, modest effect | A few very small Japanese trials | Very limited |
| Most studied situation | Older adults, mild insomnia | Restricted sleep, self-reported quality | Insomnia complaints, only one notable trial |
| Studied dose | 250–500 mg of elemental magnesium daily | 3 g before bed | 300 mg daily |
| Main side effect | Diarrhea, drug interactions | Rare digestive problems | Poorly documented |
| My choice for a first trial | If dietary intake is low | If sleep feels unrefreshing | Not first choice |
These are study doses, not recommendations. For magnesium, EFSA sets a safety benchmark of 250 mg daily for added magnesium (supplements, fortified waters), excluding food.
Why I advise against starting with a “stack”
Taking all three at once is the best way to learn nothing. If you sleep better, you will not know which one works. If you have digestive problems, you will not know which one causes them. And you are paying for three products to test one hypothesis. It is the start of the Pill Cupboard: bottles piling up with no one knowing which does what.
A better approach:
- Track your sleep for a week without supplements: bedtime, estimated time to fall asleep, awakenings, and how you feel on waking out of 10.
- Choose just one supplement based on the table.
- Take it for two to four weeks, at the same time, without changing anything else.
- Compare with your baseline week. If there is no clear difference, stop rather than adding a second one.
During this trial, the free measures still come first: a fixed wake-up time, morning light, cutting off caffeine early, and a cool bedroom. They carry more weight than any capsule. If you take regular medication, are pregnant or have kidney disease, ask your doctor or pharmacist before starting.
