UNFAIR
Download
Blog · Podcasts

The Andrew Huberman Supplement Stack (2026)

Andrew Huberman's supplement list has become internet canon, but the public stack keeps changing. This review sorts the compounds with real evidence, the occasional tools, the situational add-ons, and the ones that should not be copied without labs.

Last updatedApr 21, 2026ByUnfair TeamRead17 min
This content is for informational purposes only and is not a substitute for professional advice.

Andrew Huberman's public supplement list gets flattened online into one giant stack, which makes the whole thing look more settled and more transferable than it actually is. His public comments across Huberman Lab describe a mix of daily habits, occasional tools, older experiments, and compounds he discusses without any good reason for the average reader to copy. The only useful question is which parts of that list still make sense once the strong evidence is separated from the podcast-era enthusiasm.

Once that filter is applied, the picture shrinks fast. The best parts of the stack are familiar: creatine, omega-3 fatty acids, correcting a real vitamin D gap, and in some cases magnesium. The weaker parts are also familiar: testosterone-oriented add-ons with thinner human data than their reputation implies, nootropics that work better as rare tools than daily habits, and longevity compounds that still outrun the human evidence. This post keeps the hormone, peptide, and longevity pharmacology territory intentionally out of scope — those decisions belong inside a physician relationship, not a blog post — and focuses on the supplement decisions a careful biohacker can act on today.

The operating principles that hold across the full list:

  • [Creatine](/library/creatine) is the easiest part of the stack to defend. Best evidence, lowest friction, clearest training upside for active readers. If you copy one compound first, start here.
  • [Omega-3](/library/omega-3) deserves earlier placement than most nootropics. Huberman has repeatedly argued for EPA intake for brain and cardiovascular support, and that advice maps onto a real food-intake gap for anyone who rarely eats fish.
  • [Vitamin D](/library/vitamin-d) belongs in the bloodwork bucket, not the blind-stacking bucket. Deficiency correction after a lab is sensible. Perpetual high-dose use without testing is not.
  • The sleep stack spans a wide range of evidence. Magnesium has a more defensible place than apigenin, GABA, glycine, or inositol, and even magnesium works best when sleep behavior is already in order.
  • Hormone-oriented compounds should not be treated as a plug-and-play upgrade. Tongkat Ali, Fadogia, boron, and related compounds belong on the other side of a lab draw and a clinician relationship. They are out of scope for this post.
  • Alpha-GPC is a real focus and training tool, not a daily staple. Huberman's framing — occasional use for hard sessions — is the right one.
  • The full stack only makes sense inside a measured system. Readers who are measuring nothing should shrink the list aggressively and treat any additional compound as a short, bounded experiment with a stop rule.

The full public stack

The table below summarizes the compounds Huberman has publicly discussed taking or trialing across several years of Huberman Lab episodes. Treat it as a map of the territory, not a prescription. The status column reflects how this rollup rates each compound for a reader who is not Andrew Huberman. Unfair's evidence-first prioritization uses the same filter.

CompoundHuberman-cited contextTypical dose discussedStatus for a general reader
CreatineStrength, power, cognition~5 g/dayStart here
Omega-3 (EPA + DHA)Brain and cardiovascular support~1.5–3 g EPA/dayStart here, calibrated to Omega-3 Index
Vitamin DDeficiency correctionBloodwork-guidedStart here only with labs
Magnesium threonateSleep support~145 mgStart here if sleep or intake gap is plausible
L-theanineSleep support, downshifting100–400 mgOccasional tool
ApigeninSleep support50 mgOccasional tool; see skip-if below
GlycineSleep support~2 gOccasional tool
GABASleep support100 mg–1 gOccasional tool
InositolSleep maintenance~900 mgOccasional tool
Alpha-GPCFocus, hard training~300 mgOccasional tool
L-tyrosineShort-term focus~500 mgOccasional tool
PhenylethylamineShort-term focus~500 mgOccasional tool
CaffeineAlertness, training outputVariableFoundation when timing is respected
ZincDeficiency correction, immune11 mg (RDA for men)Deficiency-correction only
Tongkat AliTestosterone-oriented~400 mg/dayOut of scope for this post; clinician-led
Fadogia agrestisTestosterone-oriented~400–600 mg, often cycledOut of scope for this post; thin human evidence
BoronTestosterone-oriented2–4 mg/dayOut of scope for this post
AshwagandhaStress, cortisol250–600 mg dividedSituational; rotate with caution
Rhodiola roseaFatigue, performanceModest capsule dosingSituational
AG1 / greens powderMicronutrient insuranceOne serving dailyConvenience, not required
MultivitaminHabit-based backstopOnce dailyGap-closer at best
Digestive enzymesDigestive supportMeal-timedSymptom-driven
GingerDigestive supportMeal-timedSymptom-driven
Grape seed extractVascular support400–800 mgLower-priority add-on
NMNLongevity, cellular energyGram-levelExperimental; thin human data
NRLongevity, cellular energy~500 mgExperimental; thin human data
NACRespiratory, illness~600 mgSituational
GlutamineGut, recoverySeveral g/daySituational

That is 27 compounds. The defensible core is four or five. Everything between those two numbers depends on labs, training context, current gaps, and whether outcomes are being tracked with a real feedback loop.


Start here — the short list

Huberman's supplement commentary works best as a hierarchy. Food quality, protein adequacy, sleep timing, exercise, and light exposure come first. The supplement layer starts after those. Once the foundation is in place, only a few compounds survive as broadly reasonable choices for a general reader.

The short list is creatine, omega-3, a bloodwork-guided decision on vitamin D, and — for some people — magnesium. These four have the strongest evidence, the lowest risk, and the least need for personalization.

Creatine

The most common context Huberman uses for creatine is the Nutrients for Brain Health solo episode.

Huberman Lab Transcript

What Huberman has said. Creatine is one of the clearest recurring supplements in Huberman's public stack. In Nutrients for Brain Health and Performance, he describes it as useful for muscle performance and for brain function, and says the cognitive literature appears to center around roughly five grams per day. The framing lines up with how he has talked about training and recovery elsewhere on the show: keep supplements boring and pick the few that survive real scrutiny.

What the evidence says. Creatine is the easiest "yes" in sports nutrition. Recent meta-analyses continue to show support for strength, power, and lean-mass outcomes when creatine is paired with resistance training (Chilibeck 2024, Nunes 2025). The brain-health story is less settled than the muscle story but is at least plausible enough that Huberman's emphasis is easy to defend. The upside is clear, the cost is low, and the risk profile in healthy adults is far better than its reputation.

Dose and form.

DoseFormContext
3–5 g/dayCreatine monohydrateBest default for most active adults
20 g/day for 5–7 days, then 3–5 g/dayMonohydrate loading phaseFaster saturation if the effect is wanted sooner
3–5 g/dayMonohydrate, any time of dayDaily consistency matters more than exact timing

Skip if. Skip casual self-prescribing if you have known kidney disease, complicated renal monitoring, pregnancy-related questions, or a medication pattern that already makes fluid balance and kidney markers hard to interpret. For healthy adults, the bigger mistake is usually overthinking it.

Omega-3

The omega-3 discussion Huberman returns to most often also appears in the Nutrients for Brain Health episode and in the 2026 Rhonda Patrick conversation.

Huberman Lab Transcript

What Huberman has said. Huberman has been unusually consistent on omega-3s. He argues that most people should aim for at least about 1.5 grams of EPA per day and says his own intake often lands around 2 to 3 grams of EPA when he is not eating much fish. Of all the supplements on his list, omega-3 may be the one that most cleanly matches a common real-world intake gap.

What the evidence says. The case for omega-3 starts with intake status. If you rarely eat fatty fish, the odds of being below target are good. The NIH Office of Dietary Supplements keeps the guidance grounded: fish and seafood remain the preferred source, and supplements are a practical way to raise intake when food patterns are not doing the job. Huberman's framing overlaps with Rhonda Patrick's in an important way: both lean toward measuring status (Omega-3 Index) rather than treating fish oil like a ritual.

Dose and form.

DoseFormContext
~1.5 g EPA/dayFish intake or supplementHuberman's lower-end target for many adults
~2–3 g EPA/dayFish oil or algae-derived omega-3His more aggressive personal intake when fish is low
Food-firstFatty fish (salmon, sardines, herring, mackerel)Best first move if you already eat fish regularly

Skip if. Skip blind high-dose supplementation if you already eat a lot of fatty fish, if you are near a procedure where bleeding-risk questions matter, or if you are on anticoagulants and have not discussed the dose with a clinician. Quality and oxidation matter more than marketing language here; see Unfair's third-party testing guide.

Vitamin D

What Huberman has said. Huberman has described taking vitamin D in the past, including relatively high doses in podcast conversations, and has also spoken about regular blood work. The blood work matters more. A dose only makes sense once it is tied to an actual level and a repeat test.

What the evidence says. Vitamin D is where a reasonable correction habit often turns into a cargo-cult stack habit. The NIH Office of Dietary Supplements lays out the basics clearly: status varies with sun exposure, latitude, age, skin pigmentation, and body composition, and the tolerable upper intake level for adults is 4,000 IU per day unless you are under clinical supervision. Doses above that are sometimes appropriate under medical guidance, but the average reader should get a lab value before borrowing a podcast dose.

Dose and form.

DoseFormContext
Lab-guidedVitamin D3Best default approach
Maintenance after correctionVitamin D3Only after a measured low level has been corrected
Food and sun exposure firstDiet plus sun where appropriateUseful background, rarely enough by itself in high-risk groups

Skip if. Skip copycat megadosing if you do not know your 25-hydroxyvitamin D level. Skip casual stacking if you already take a multivitamin, fortified foods, or a clinician-directed vitamin D protocol and have not added up the total intake.

Magnesium and the sleep stack

What Huberman has said. Huberman has returned to the same sleep stack for years: magnesium threonate, apigenin, and theanine, with glycine, GABA, and inositol discussed as additional tools in some contexts. The clearest transcript reference is his Sleep Toolkit episode, where he mentions about 145 mg of magnesium threonate, 50 mg of apigenin, and 100 to 400 mg of theanine, with explicit caveats that some people do poorly with parts of the stack. The warnings are easy to miss because the internet usually copies the ingredients and drops the caveats.

What the evidence says. The evidence behind these compounds varies widely. Magnesium has the most defensible place, especially when dietary intake is low or deficiency is plausible (NIH ODS magnesium fact sheet, Pitre 2021 meta-analysis). The magnesium-and-sleep evidence is mixed rather than airtight, with some positive trials and some low-certainty reviews. L-theanine has a small but growing sleep literature. Apigenin remains more speculative than its popularity suggests and Huberman himself has flagged estrogen-suppression concerns in later sleep episodes. Glycine, GABA, and inositol are best treated as occasional tools rather than universal nightly habits.

The structural point matters here too: when sleep timing, caffeine timing, alcohol, late light exposure, and bedroom conditions are still poor, any sleep stack ends up doing work that behavior should be doing first. Fix the habits, then evaluate the supplements.

Dose and form.

DoseFormContext
~145 mgMagnesium threonateHuberman's often-cited sleep-stack dose
100–400 mgL-theanineSleep aid for people who tolerate it well
50 mgApigeninExperimental add-on, not a foundational deficiency correction
~2 gGlycineOccasional use, not a universal habit

Skip if. Skip the full stack if sleepwalking, night terrors, vivid-dream disruption, GI intolerance, or hormone-related concerns already make the trade-off unattractive. Apigenin especially does not deserve default use in anyone who has not thought through the estrogen-suppression concern Huberman has flagged. And anyone on psychoactive medication should talk to a clinician before layering GABA-ergic compounds on top of a prescription.


Occasional tools — focus and training support

This tier is the supplements Huberman has publicly used as situational tools rather than daily staples. The utility is real; the "every single morning" framing is not.

Alpha-GPC

What Huberman has said. Huberman has discussed alpha-GPC as a choline donor for focus work and hard training sessions. In Nutrients for Brain Health he describes using it at about 300 mg from time to time rather than as a daily staple. The occasional-use framing is what makes it a real tool — pulled out for specific sessions rather than taken every morning.

What the evidence says. Alpha-GPC works as a choline donor that supports acetylcholine production, and there is evidence for both cognitive support and power output in training contexts. One published concern worth noting: a 2021 observational cohort study in JAMA Neurology found an association between alpha-GPC use and higher 10-year stroke incidence in a Korean population. The study found a correlation, not a causal link, and the population, dose context, and confounders make it hard to generalize to Western supplement use at typical doses. Worth knowing about, especially for anyone with existing cardiovascular risk, but a single observational finding does not settle the question. For a related compound with more cognitive-enhancement literature, see CDP-choline.

Dose and form.

DoseFormContext
~300 mgAlpha-GPCOccasional use for focus sessions or hard training

Skip if. Skip this compound as a daily habit. Anyone with cardiovascular risk factors should start conservatively and track carefully, or choose CDP-choline instead.

L-tyrosine, phenylethylamine, and caffeine

What Huberman has said. Huberman has mentioned L-tyrosine (~500 mg) and phenylethylamine (~500 mg) as short-term focus aids. Caffeine is the one stimulant he discusses with real timing discipline — morning rather than afternoon, with a delay after waking to let adenosine clear.

What the evidence says. L-tyrosine and phenylethylamine are better understood as short-term arousal tools with variable tolerance and obvious downside risk if sleep, anxiety, or stimulant load are already poor. Caffeine is the most defensible stimulant on this list because the trade-off is well understood. The downside is also obvious, which is why caffeine timing should usually be fixed before any exotic nootropic gets added. See Unfair's cognitive performance and nootropic stacking post for the broader framing.

Dose and form.

DoseFormContext
~500 mgL-tyrosineShort-term focus or late-work push, not a default habit
~500 mgPhenylethylamineOccasional stimulant-style use with a short time horizon
VariableCaffeineTiming discipline matters more than the dose

Skip if. Skip this bucket if anxiety, blood-pressure issues, stimulant sensitivity, poor sleep, or a heavy caffeine habit already describe your baseline.


Hormone-oriented compounds — out of scope for this post

Tongkat Ali, Fadogia agrestis, and boron are the most internet-famous pieces of Huberman's public list, which is also why they deserve the most careful handling. Huberman himself has always framed these inside a bloodwork-and-clinician context. Every time they get flattened into a plug-and-play stack, something important gets lost.

For this rollup specifically, the hormone-oriented compounds are out of scope. They are not out of scope because they are taboo; they are out of scope because any useful answer depends on individual labs, medication patterns, fertility goals, cardiovascular baseline, and clinician relationship — none of which a blog post can provide. Anyone weighing these compounds should read Unfair's clinician consult guidance, confirm a baseline, and work through a medical relationship rather than a podcast recommendation.

Zinc is a narrower case because it is a real mineral with a real deficiency risk (NIH ODS zinc fact sheet). If you are treating zinc as a general wellness add-on, do not. If you have a plausible deficiency or a known dietary gap, use it as a deficiency-correction tool at the RDA (11 mg for men, 8 mg for women) — not as a hormone hack.


Longevity, recovery, and general-wellness add-ons

The remainder of the public list is the older and less stable parts of the stack: AG1, rhodiola, ashwagandha, NMN, NR, NAC, glutamine, digestive enzymes, ginger, grape seed extract, multivitamin. Some of these have real applications in the right context. All of them deserve a tighter frame than "Huberman mentioned it, add it to the daily stack."

What the evidence says. Ashwagandha is a good example of the pattern. There is enough trial data on stress-related outcomes to keep it resurfacing, but NCCIH is explicit about the cautions: liver injury reports, drug interactions, thyroid effects, and pregnancy. It is a situational tool, not a universal daily habit; Unfair's guide to managing supplement tolerance and withdrawal is the right reading before cycling it in.

NMN and NR remain experimental longevity compounds. Human data is far thinner than the longevity rhetoric that surrounds them. They should be treated as experiments with a stop rule, not as confident basics. NAC has narrower use cases — respiratory and illness-adjacent — rather than a default daily role. Glutamine has clearer indications in sport and clinical nutrition than it does as a general everyday add-on for healthy adults who already eat enough protein. Rhodiola may help some people with fatigue; it belongs in the situational-tool category, not the foundation.

Digestive enzymes and ginger are symptom-driven rather than default-use. Grape seed extract has some cardiovascular literature but still sits far below creatine and omega-3 on any priority list (NCCIH grape seed extract). A multivitamin can make sense when diet quality is erratic or when energy restriction is high; once food intake is strong and specific gaps are being tracked, the case weakens. AG1 is a convenience product, not a supplement that needs to be defended on its own merits.

Skip if. Skip this whole bucket as a first move if protein, calories, fiber, hydration, and meal structure are still disorganized. General-wellness supplements are easy ways to feel productive without fixing the higher-return layer. The usefulness-to-complexity ratio drops sharply here; minimum viable stack thinking applies.


Building your own stack from Huberman's

The full public list runs to 27 compounds. The defensible core is four or five. Everything between depends on labs, training, current gaps, and whether outcomes are actually being tracked. Sort the list into three tiers and the decision gets much simpler.

Start here. Creatine first if you lift, sprint, or care about preserving training quality. Omega-3 next if fish intake is low. Vitamin D only with bloodwork and a retest. Magnesium if sleep or intake patterns make the case plausible.

Situational, with a clear use case. These earn a slot when there is a defined reason to take them — hard training session, stress window, specific gap — and get logged as time-bounded experiments.

  • Alpha-GPC — ~300 mg before hard training or focus sessions
  • L-theanine — 100–400 mg in the evening
  • L-tyrosine — ~500 mg before a focus block, not daily
  • Ashwagandha — 250–600 mg divided, rotated not continuous
  • Rhodiola — modest capsule dosing during fatigue windows

Clinician-led or skip. These compounds either need labs and medical guidance or sit below the evidence bar for a general reader.

  • Tongkat Ali, Fadogia agrestis, boron — clinician-led only; not covered here
  • Apigenin as a nightly habit — situational at best, skip if estrogen-related concerns apply
  • NMN and NR — experimental, not high-confidence
  • Glutamine as a general add-on — situational
  • Phenylethylamine — situational stimulant use only
  • The full six-compound sleep stack as a bundle — overkill when sleep behavior is the real issue

A well-chosen stack of four to six compounds, guided by labs and tracked outcomes, will outperform a blind copy of all 27 almost every time. This is the shape of a minimum viable stack drawn directly from Huberman's public comments.


How Unfair uses this rollup

Unfair treats the Huberman list the way it treats any supplement recommendation: sort by evidence, surface dose windows, flag overlap before a new addition duplicates something already in the stack, and bias toward the short list that has actually cleared the evidence bar. The app's recommendation engine will not suggest the full twenty-seven compound list, and it will not pretend that "Huberman mentioned it" is evidence on its own. It is a data point, not a verdict.

For the episode-by-episode view of Huberman Lab's actual content, see the Huberman Lab rollup. For an adjacent perspective on many of the same supplements from a more research-forward angle, see the Rhonda Patrick rollup. For the longevity framework that sits underneath most of these decisions, see the Peter Attia rollup.