Magnesium Glycinate vs Citrate vs L-Threonate: Which Type Do You Actually Need?
I take magnesium every night. MagTech, as it happens, which we stock. I sleep better on it than off it and I have done for years, and a good number of our customers buy the same thing again and again and tell me the same story.
And I am probably the last person you should trust on this.
I will come back to why. First, the single most useful thing anyone can tell you about magnesium, which is that the cheapest and most common form of it barely works at all.
The form that does nothing
Magnesium oxide is what you will find in most supermarket magnesium, most multivitamins, and most of the cheap tablets online. In a randomised, double-blind, placebo-controlled trial of 46 adults over 60 days, the verdict on it was flat: "Mg oxide supplementation resulted in no differences compared to placebo" (Walker et al., Magnesium Research, 2003). An earlier analysis of commercial preparations put its absorption at around 4% (Firoz and Graber, 2001).
The chemistry is not complicated. Magnesium oxide is close to insoluble in water. If it will not dissolve, it will not absorb, and most of it goes straight through you. Which is precisely why the same compound is sold as a laxative. That is not an unfortunate side effect. That is the entire mechanism.
So if you have taken magnesium before and felt absolutely nothing, do not write off magnesium. Check the label first. There is a decent chance you were taking the form that does not get in.
The three that do
Citrate is the workhorse - well absorbed, well studied, cheap enough. In that same 2003 trial it produced the biggest rise in blood magnesium of anything tested. Honest caveat: it is still mildly osmotic, so at higher doses it can loosen things up.
Glycinate is magnesium bound to glycine. It is what people reach for when other forms upset their stomach, and it has a reputation for being the gentle one.
That reputation is barely tested. There is essentially one human study comparing glycinate to oxide, from 1994, in twelve patients recovering from bowel surgery, and it found no absorption advantage whatsoever (Schuette et al., 1994). The bit about better tolerance was an offhand observation by the researchers, not something they measured. Nobody has ever run a proper trial on it in healthy people.
What we can say is that badly absorbed magnesium sits in the gut, pulls water in after it, and that is what causes the trouble. Better absorbed forms leave less behind. So "glycinate is gentler" is a reasonable inference from the absorption data, and an awful lot of people swear it is true. It is just not a finding, and I am not going to dress it up as one.
L-threonate is the newest, the dearest, and the one with the brain claims on the box. We will get to that.
And the sleep question?
It is why half of you are here, and it deserves - and now has - its own full article. The short version: there is no EU-authorised claim linking magnesium to sleep, the trials are thin, and the best independent one found an improvement smaller than you would actually notice. And yet - magnesium shortfall is real and common, it wrecks sleep, and if you are one of the people running low, correcting it can change things noticeably. Trials recruit whoever turns up, most of whom are not short of magnesium, so a strong effect in the people who needed it gets flattened into a weak average across everyone else. That may be you. It may not.
The full honest argument, including why I do not trust my own good nights on the stuff, is in the sleep piece. This article's job is the other question: which form to buy.
What magnesium is actually allowed to claim
These are the effects assessed and permitted under EU law. Not marketing lines - the real list:
- Normal muscle function
- Normal functioning of the nervous system
- Normal psychological function
- Reduction of tiredness and fatigue
- Normal energy-yielding metabolism
- Normal protein synthesis
- Electrolyte balance
- Maintenance of normal bones and teeth
Nervous system, psychological function, tiredness. All there. Sleep, nowhere. Make of that what you will.
And the brain claims?
L-threonate is sold almost entirely on cognition. The research is genuinely interesting and nowhere near as settled as the packaging suggests.
The study everybody cites ran in Neuron in 2010 and it is good science. It is also a study in rats (Slutsky et al., 2010). The one human trial, from 2016, had 44 participants who already had cognitive complaints, was run by the company that makes the ingredient, has never been independently replicated, and could not determine any effect on sleep or anxiety at all because the placebo response drowned it out (Liu et al., 2016).
A live line of research, then. Not a proven benefit. There is no authorised EU claim connecting magnesium to memory, and I am not about to invent one.
Where we landed
We stock two magnesiums these days, and they are for different people.
MagTech is the serious capsule - L-threonate, taurate and glycinate, no oxide. The logic is simply that if you are going to take magnesium at all, take the forms that get in and stop paying for the one that does not. That is the entire pitch, and it is the one I take.
Zen Magnesium Gummies are the easier one - 200mg of magnesium with ashwagandha and L-theanine, green apple flavour, no capsule to swallow. If the thing stopping you taking magnesium has always been remembering to take magnesium, a gummy you enjoy solves more than it looks like it should.
And if you would rather buy a plain citrate from someone else for half the price, that is a perfectly sound decision, and you now know enough to make it well. I would rather you bought the right thing than bought from me.
As for whether it will fix your sleep - that is the other article.
Frequently asked questions
Does magnesium help you sleep?
There is no EU-authorised claim linking the two, the trial evidence is rated low to very low quality, and the people likeliest to notice a real difference are those whose intake was low to begin with. The full honest answer - trials, caveats and all - is in our magnesium and sleep article.
Which magnesium is best absorbed?
Citrate has the strongest evidence among the common forms. Glycinate and L-threonate are also well absorbed. Magnesium oxide is poorly absorbed, at roughly 4%, and in a randomised trial did no better than placebo.
Is magnesium glycinate really gentler on the stomach?
Probably. It is not proven. The only human comparison is a 1994 study in twelve post-surgical patients which did not formally measure digestive symptoms. The reasoning is sound, the reports are widespread, the evidence is thin.
Does magnesium L-threonate improve memory?
The headline research is in rats. The single human trial had 44 people and was run by the ingredient's manufacturer. Interesting, unproven, and not an authorised claim.
When should I take it?
Whenever you will actually remember to. Consistency beats cleverness, every time.
Food supplements are not a substitute for a varied, balanced diet and a healthy lifestyle. If you are pregnant, breastfeeding, on medication, or have a kidney condition, have a word with your GP first. And if your sleep has been genuinely bad for a long time, that is a conversation for your doctor, not a jar of capsules.