Writing — Comparisons

Threonate vs Glycinate: Which Magnesium Form Is Actually Worth Your Attention

Every supplement is a small work of fiction until someone actually measures it. Magnesium is no exception. If you have spent any time near the supplement aisle lately, you have met the two most popular contenders: magnesium L-threonate and magnesium glycinate. One is marketed as a brain-and-cognition supplement. The other is positioned as the sensible way to fix a deficiency. They share a single mineral and almost nothing else — and the people charging you a premium for the one marked "brain" would prefer you did not read this far.

So let us leave the campaign copy at the door and do the boring, useful thing: what each form actually is, what the evidence does and does not support, and which one you should take. SCRUPLE does not ship magnesium. We raise it the way we raise every topical claim — because a serious stack should be built from evidence, not swallowed from a label.

What magnesium threonate actually is

Magnesium L-threonate is magnesium bound to L-threonate, and the threonate half is not decoration: preclinical work indicates it is carried by glucose transporters, which is the proposed reason this compound raises magnesium inside the brain rather than just in serum [4]. It arrives in the literature with a 2010 paper in Neuron, which makes it young by supplement standards [1]. Young is not a synonym for bad. It is a synonym for "only recently put through clinical testing."

The thesis behind it is straightforward and defensible: this particular form appears able to raise magnesium levels in the brain, and raising brain levels may support memory, learning, attention, and sleep. Note the cautious wording. That sentence is the difference between what the brand says and what the evidence says. The origin study is a rat study — magnesium L-threonate increased synapse density in the hippocampus and improved learning and working memory in rats, including pattern completion in aged rats (Slutsky et al., Neuron, 2010) [1]. Rats are a reason to run a human trial. They are not a human result.

Then the human trials, all three of which are worth naming because the pattern is the point:

• 51 adults aged 50–70 with self-reported memory complaints, 12 weeks. Composite cognitive ability improved against placebo, and the effect was large — Cohen's d = 0.91, 95% CI 0.27–1.51, p = 0.003. Funded by Neurocentria, which was developing the compound (Liu et al., Journal of Alzheimer's Disease, 2016) [2].
• 109 healthy Chinese adults aged 18–65, 30 days. All five subtests of the Clinical Memory Test improved against placebo, p < 0.001, with the memory quotient rising from 60.3 to 81.8. Funded by Wuxi Magtein Bioscience (Zhang et al., Nutrients, 2022) [3].
• 100 healthy adults aged 18–45 with poor sleep, six weeks. The total cognition composite improved by 8.40 points versus 5.60 on placebo, p = 0.043, and sleep-related impairment improved, p = 0.043 — but there was no difference in Raven's 2 reasoning, p = 0.953, and no difference in any of the Oura Ring sleep measures. Funded by Threotech, which supplied the product (Lopresti and Smith, Frontiers in Nutrition, 2026) [4].

Read those three lines again. The samples are 51, 109 and 100. Every one of them was paid for by a party with a commercial interest in the answer. And the most recent trial — the one in younger, healthier people — is also the one where the wins get narrower and the objective sleep tracking shows nothing. Its own authors write that the wider threonate literature is "negatively impacted by open-label designs, small sample sizes, and, in most cases, the delivery of Magtein in combination with additional nutrients that could also affect cognitive function" [4]. Small groups, funded designs, and a plausible mechanism are not proof. They are a reason to be interested. The confidence you can reasonably place in threonate's cognition claims is lower than the confidence the marketing assumes.

What magnesium glycinate is

Magnesium glycinate is the mineral bound to glycine, an amino acid your body already uses heavily. It is a gentle, absorbable form, and that handles the two problems people discover magnesium for: low levels and an unsettled stomach. Solubility is the mechanism there — forms that dissolve well are absorbed more completely, and the forms most often reported to cause diarrhoea are carbonate, chloride, gluconate and oxide (NIH Office of Dietary Supplements) [9]. Head-to-head bioavailability testing of commercial magnesium products confirms the pattern is driven by how well the formulation dissolves, not by the name on the front of the bottle (Blancquaert et al., Nutrients, 2019) [10].

The evidence pool behind glycinate is older and broader, and the honest word for the size of the effects is "modest." Here is what modest means in numbers.

Blood pressure. A meta-analysis of 34 randomised, double-blind, placebo-controlled trials in 2,028 adults found magnesium supplementation lowered systolic pressure by 2.00 mmHg, 95% CI 0.43–3.58, and diastolic by 1.78 mmHg, 95% CI 0.73–2.82. The median dose was 368 mg a day for three months (Zhang et al., Hypertension, 2016) [6]. Real, replicated, and about two points.

Blood sugar. A systematic review and meta-analysis of randomised trials found magnesium improved insulin resistance — HOMA-IR weighted mean difference −0.67, 95% CI −1.20 to −0.14, p = 0.013 — while the effect on fasting plasma glucose across all trials did not reach significance, p = 0.119, and only appeared in trials running four months or longer (Simental-Mendía et al., Pharmacological Research, 2016) [7]. Note the time horizon. Four months, not four nights.

Sleep. This is the reason most people come to this article, and it is the weakest of the three. A systematic review of nine studies in 7,582 adults found observational evidence linking magnesium status to sleep quality, but reported that the randomised trials contradict each other and the relationship between supplementation and sleep disorders remains uncertain (Arab et al., Biological Trace Element Research, 2023) [8]. The best recent trial on glycinate specifically — 155 adults with poor sleep, four weeks, 250 mg of elemental magnesium a day as bisglycinate, and notably not funded by a supplement company — did beat placebo on the Insomnia Severity Index: −3.9 points against −2.3, p = 0.049. The authors report the effect size as Cohen's d = 0.2, which is small, and the paper says so (Schuster et al., Nature and Science of Sleep, 2025) [5].

The caveat that matters: most of that research is on magnesium as a mineral across many forms, at full useful doses. Glycinate inherits that wider evidence base. It works, not because it is magic, but because it is magnesium.

The no-marketing question: what are you actually trying to fix?

Every comparison post collapses into a question the label will not ask you: what is the target. The answer decides everything.

• If the goal is to correct a likely deficiency — whether you are an athlete, on a prescribed medicine, young, or simply someone who has eaten poorly for a decade — then plain magnesium does that, and glycinate is a fine form for it. It is not glamorous. It is effective. Take a full elemental dose.
• If the goal is cognitive support and sleep because you run on caffeine and want the edges off the jitter, then threonate is the hypothesis worth testing: raising brain levels specifically. It is a premium bet, and it may or may not hold on you.

Two forms, two different dose maths

Read the label for the elemental magnesium content, not the total weight of the compound. This is where the whole argument gets settled, and it gets settled by arithmetic rather than opinion.

The recommended dietary allowance is 400–420 mg a day for adult men and 310–320 mg for adult women, most of which is meant to come from food. The tolerable upper intake level for magnesium from supplements specifically is 350 mg a day (NIH Office of Dietary Supplements) [9]. Now put the two products next to those numbers.

Two grams of magnesium L-threonate — the full dose used in the trial — delivers 145 mg of elemental magnesium [4]. That is roughly 7 percent of the compound by weight, and it is under half the RDA. It was never designed to correct a deficiency, and taken alone it will not. The bisglycinate sleep trial ran at 250 mg of elemental magnesium [5], and the blood-pressure trials clustered around 368 mg [6].

So the part nobody wants to say: the higher dose is on the glycinate side, and the higher price is on the threonate side. When a form is both less proven and delivers a fraction of the elemental magnesium per gram, the value proposition starts to look a little lopsided. Not wrong — just lopsided, and knowable from the label before you pay.

What they both can actually do for sleep

Here is the honest overlap. Both forms carry magnesium into your biology, and magnesium acts on the nervous system in part by blocking the NMDA receptor channel — the rat study behind threonate reports upregulated NR2B-containing NMDA receptors and enhanced long-term potentiation as the mechanism for its memory effects [1]. Both forms can, in practice, help with sleep, and neither has strong trial evidence that it does so reliably: the systematic review says the randomised trials contradict each other [8], and the best glycinate trial found a small effect [5]. If your concern is simply the evening brightness of your own brain, then form matters less than you have been told. This is not hedging. It is what the papers say.

The higher dose is on the glycinate side. The higher price is on the threonate side.

The contrary answer, in three sentences

If you are an evidence-literate adult who builds stacks and chases focus, here is our actual recommendation.

Glycinate is the honest generalist. If you suspect you are low, or just want the mineral covered at a dose the research actually uses, take a full elemental dose of glycinate. It is gentle, cheap, and evidence-backed.

Threonate is the specialist hypothesis. If your target is specifically a brain-forward, sleep-and-cognition-friendly form from a smaller, newer evidence base, and you are comfortable paying a premium for a more speculative claim, threonate is the bet you picked. It can be fine. Just know the premium for what it is.

Do not stack both to hedge. The upper limit for supplemental magnesium is 350 mg a day, and high supplemental doses commonly produce diarrhoea, nausea and abdominal cramping (NIH Office of Dietary Supplements) [9]. A working dose of glycinate plus threonate is how you end up with an expensive trip to the restroom. Pick one, dose it honestly, and wait four months before you decide anything [7].

And neither form is a substitute for the thing that actually got you here: caffeine and a keyboard. Magnesium is not a focus stimulant. It will not replace the active edge you came looking for. The measured approach for a caffeine drinker is to build around the caffeine itself — pair it with an amino acid at a ratio that has actually been trialled, such as the 100 mg of L-theanine with 50 mg of caffeine that improved attention-switching speed and accuracy at 60 minutes (Owen et al., Nutritional Neuroscience, 2008) [11], or the 200 mg with 160 mg that improved hit rate and hazard reaction time in sleep-deprived adults (Nawarathna et al., British Journal of Nutrition, 2025) [12] — scale that to your body weight, and treat magnesium as a separate decision made for its own reasons.

The whole argument

We end where SCRUPLE always ends: your dose is not a serving suggestion.

The claim gets stated with confidence, then a label names a serving. The dose math, the elemental magnesium, the time horizon of weeks — those go in the footnote. So the real question, threonate or glycinate, is not which form fixes your sleep. It is why you are taking either, and at what dose, and on what evidence, and how long you will give it before you decide anything.

The magnesium shelf is a perfectly useful destination. It is not the biggest decision you will make this month. Both forms work, both can run overpriced, and neither earns the religion people assign to it. Choose for the reason and the dose, not the slogan. That is the whole argument.