Magnesium glycinate vs citrate: which should you actually take?

It is 11pm, you are three tabs deep into supplement advice, and every tab agrees you should be taking magnesium. They just cannot agree on which one. Glycinate for sleep, says one. Citrate for absorption, says another. Threonate for your brain, whispers the expensive one. Same mineral, six suffixes, and a price range that runs from pocket change to boutique skincare.

Here is the fact none of those tabs mention, and it is the strangest thing we found researching this piece: the internet's favorite supplement matchup has never actually been run. There is no human trial that puts glycinate and citrate head to head. Thousands of articles are refereeing a fight that never happened.

So let's work with what has actually been tested, because there are real answers here. They are just not where the labels point.

Glycinate vs citrate: the fight with no referee

The closest science gets is a small 2003 trial that compared citrate, oxide, and an amino-acid chelate, which is the family glycinate belongs to. Around ten people per group, sixty days. Citrate and the chelate both clearly beat oxide, and citrate alone nudged blood and saliva magnesium up. The authors themselves called the citrate-versus-chelate comparison preliminary and asked for a proper study. Twenty years later, nobody has run one.

So the honest scorecard reads: both forms absorb well, citrate holds slightly better direct evidence, and glycinate's famous gentle-on-the-stomach reputation rests mostly on a 1994 study showing chelated magnesium can ride into your body through a peptide doorway. Plausible chemistry, thin receipts. One practical difference is real, though: citrate pulls water into the bowel. If you run constipated, that is a feature, and medicine has used it as a mild laxative for decades. If you do not, glycinate is the reasonable bet.

The form to actually avoid

While the internet argues about the top two, the form in most of the cheapest bottles quietly loses every study it enters. Magnesium oxide showed fractional absorption of about 4 percent in the small study that measured it, and in the 2003 trial it performed no differently from placebo. The NIH lists it as a laxative ingredient, which is a polite way of saying most of it passes through you with a mission.

It gets worse. In 2025, third-party testing by ConsumerLab flagged two products that appeared to contain cheap oxide instead of the glycinate and aspartate on their labels. One company admitted it. You were not just buying the weakest form. Sometimes you were buying it dressed up as the strong one.

Does magnesium actually help you sleep?

Modestly, maybe, on evidence that would embarrass a better-funded field. The meta-analysis that everyone cites found magnesium helped older adults fall asleep about 17 minutes faster than placebo. That number comes from three trials totalling 151 people, which the reviewers themselves graded low to very low quality. Anxiety is a similar story: a systematic review found the evidence suggestive but poor, and a newer depression meta-analysis found a real-looking effect across seven small, messy trials.

Our read: seventeen minutes is not nothing, magnesium is cheap, and the side-effect bar is low. But anyone telling you it is proven sleep medicine is quoting confidence the studies never earned.

Is threonate worth three times the price?

Magnesium L-threonate is the premium option marketed for memory and brain fog, and its origin story is genuinely cool: it improved learning and memory in rats in a 2010 study. The human evidence since then has a pattern worth knowing before you pay for it. The main cognition trial had 44 people and was funded, designed and written up by the company commercializing the compound. The 2024 sleep trial was funded by the ingredient's distributor. The 2025 trial was funded by another seller. Every human study we could find was paid for by someone selling threonate. That does not make the results fake. It makes them unconfirmed, at three times the price of forms with independent evidence.

The label trick to check before you buy

Now the part that got companies in legal trouble this year. Magnesium doses can be written two ways: the weight of the whole compound, or the elemental magnesium your body actually gets. A bottle shouting magnesium glycinate 1000 mg may deliver somewhere near 100 to 140 mg of actual magnesium. By law, the Supplement Facts panel on the back must state the elemental amount. The front of the bottle is marketing. The back is the contract.

This is not hypothetical. In June 2026 a major brand agreed to pay over 1.8 million dollars to settle claims its product contained far less magnesium than listed, and two months earlier another agreed to stop implying its product delivered nine times its real dose. Third-party testing has also found the price tells you nothing: products costing twenty times more than the cheapest passing options have failed quality testing. Read the back panel. Ignore the front.

Do you even need to supplement?

Possibly not. Yes, federal survey data says about 48 percent of Americans eat less magnesium than the estimated requirement, which is the statistic every supplement ad leans on. Eating below the target is not the same as being deficient, and your body buffers hard. The frustrating twist is that testing will not settle it either: less than 1 percent of your magnesium lives in blood serum, so the standard blood test barely reflects your stores. The boring, evidence-solid move is food first. Pumpkin seeds, almonds, spinach, black beans and whole grains get most people to the target, roughly 400 to 420 mg a day for men and 310 to 320 for women, no capsule required.

Who should skip the supplement aisle entirely

Anyone with impaired kidneys, because that is where real magnesium toxicity happens, and anyone stacking doses past the supplemental safety ceiling of 350 mg a day without a clinician involved. Magnesium also blocks absorption of some antibiotics and osteoporosis drugs, so if you take those, space them by at least two hours. And skip the exotic forms sold on heart and energy claims, like taurate and malate. Their evidence is animal studies and extrapolation, which is a long way from your cardiologist's approval.

The bottom line

Constipated? Citrate, and it is doing two jobs. Sensitive stomach? Glycinate is a fair bet, just unproven as the winner. On a budget? Any citrate that lists its elemental dose beats a premium bottle that hides it. Avoid oxide unless you want the laxative. Treat threonate as an experiment you fund. And whatever you buy, flip the bottle around, because the back panel is the only part of the label that answers to regulators.

What we still do not know, said plainly: whether glycinate or citrate is actually better, because the trial has never been run, and whether threonate's promise survives independent replication. If either changes, this article changes, and you will see it logged in the updates section below.