If you have searched for the best supplements Andrew Huberman recommends, you have probably already heard the long list on the podcast: AG1, magnesium L-threonate, apigenin, ashwagandha, Rhodiola, alpha-GPC, omega-3, vitamin D3 with K2, creatine, Tongkat Ali, Fadogia agrestis, urolithin A.
Quick Answer: which Huberman supplements actually have evidence?
The 2 to 3 we would actually start with:
- Creatine monohydrate, 5 g/day. The most replicated cognitive and physical-performance supplement on the market, with hundreds of RCTs and a 2017 ISSN position stand calling it safe and effective.
- EPA-dominant omega-3 fish oil, roughly 1,000 mg EPA/day. Meta-analytic signal across mood, cognition, and cardiovascular endpoints.
- Magnesium glycinate or L-threonate, 200 to 400 mg of elemental magnesium at night. Sleep and stress signal in a population that is widely magnesium-low.
Who should not start with these: anyone already on prescription stimulants, SSRIs, or thyroid medication should add one supplement at a time, not a Huberman-style 8-bottle stack, because you cannot attribute effect to any single intervention. Anyone on warfarin or other anticoagulants needs prescriber clearance for high-dose fish oil. Men with normal testosterone considering Tongkat Ali plus Fadogia together should pause and read the trial data first.
What to do FIRST: if your problem is sleep, the bedroom environment and a consistent wake time outperform any supplement. If your problem is energy or focus, an iron panel, thyroid panel, and 25-hydroxyvitamin D test name the real bottleneck more reliably than guessing at apigenin or alpha-GPC. Supplements come after the boring work.
Where this stack actually comes from
Andrew Huberman is a Stanford professor of neurobiology and ophthalmology whose podcast has popularized a particular supplement stack across multiple long-form episodes. He has a publicly disclosed commercial partnership with Momentous, a supplement company that produces several of the items he discusses (creatine, Tongkat Ali, magnesium L-threonate, urolithin A under the Mitopure label). That partnership is relevant context, not a disqualifier. It means the brand-specific recommendations on the podcast should be read as a partnership disclosure, not as independent evidence that Momentous is the best brand for any given compound.
What the podcast does well is name compounds with a plausible mechanism and a published trial somewhere. What it does less well is signal which trials are large, replicated, and clinically meaningful versus which are small pilots or animal studies. The split below is the version a research-trained reader gets when looking only at the human RCT evidence.
None of these supplements treat a diagnosed condition in any FDA-recognized sense, and Huberman is careful about that on-mic. For specific clinical issues (insomnia disorder, major depression, hypogonadism, ADHD), the standard of care is a clinician evaluation and, where indicated, prescription medication. For a comparison stack with a longevity-driven framing, see our piece on the Peter Attia supplement stack.
Strong evidence: the parts of the stack with real human RCT support
Creatine monohydrate
Why it helps. Creatine increases intramuscular phosphocreatine stores, which buffer ATP regeneration during short, high-output efforts. The brain uses the same phosphocreatine system, and there is growing evidence that creatine modulates cerebral energy metabolism and mitochondrial function in regions like the prefrontal cortex.
What the trials show. The evidence base for performance and lean-mass outcomes is one of the deepest in supplement science, summarized in the 2017 ISSN position stand. The cognitive signal is younger and smaller: a 2018 meta-analysis found a modest effect on short-term memory and reasoning in healthy adults, with larger effects in vegetarians and the sleep-deprived.
Dose used in trials. 3 to 5 g/day for muscle outcomes; 5 g/day is the standard maintenance dose. Cognitive trials have used 5 to 20 g/day, with the larger doses showing the cleanest effects on tasks under metabolic stress.
Form to look for. Creatine monohydrate, ideally Creapure-branded for the most-studied quality grade. The ConsumerLab creatine review flagged label-claim accuracy as the main consumer issue. Skip newer fancy forms (HCl, ethyl ester, buffered) until they match monohydrate's evidence base.
Skip if you have stage 3 or worse chronic kidney disease, where the data is less clear and the prescriber should sign off.
Actionable takeaway: Creatine is the single highest-confidence pick on the entire Huberman stack and the cheapest. If you take nothing else from this article, 5 g of creatine monohydrate per day is the move.
EPA-dominant omega-3 fish oil
Why it helps. Long-chain omega-3s, especially EPA, are incorporated into neuronal membranes, modulate inflammatory eicosanoid signaling, and influence BDNF expression in animal models. The downstream effect on mood, cognition, and cardiovascular endpoints is one of the most-studied food-derived stories in nutrition.
What the trials show. A 2019 meta-analysis of omega-3 in depression found a modest but reliable effect for EPA-dominant formulations at 1 g or more per day. The cardiovascular REDUCE-IT trial used 4 g/day of icosapent ethyl (prescription-grade EPA) and reduced major adverse cardiovascular events in high-risk adults. For cognition, the Chang et al. 2017 systematic review confirmed lower baseline EPA in youth with attentional problems and modest improvement on supplementation. For brand-level picks, see our best omega-3 supplements guide.
Dose used in trials. 1,000 to 2,000 mg of EPA per day for mood and cognition.
Form to look for. Triglyceride or re-esterified triglyceride form, EPA at least 60% of total EPA+DHA, third-party tested for oxidation. Nordic Naturals, Carlson, and Wiley's Finest have repeatedly passed ConsumerLab assays.
Skip if you are on warfarin or another anticoagulant without prescriber clearance.
Dose-in-trial vs dose-people-buy. A generic 1,000 mg fish oil softgel contains roughly 180 mg EPA. To match 1,000 mg EPA, that is five to six softgels per day. Most disappointment with fish oil traces back to underdosing, not to fish oil failing.
Vitamin D3 (often paired with K2)
Why it helps. Vitamin D is a secosteroid hormone with receptors in nearly every tissue, including immune cells, neurons, and osteoblasts. Low 25-hydroxyvitamin D is associated with worse outcomes across mood, immune, and musculoskeletal endpoints in observational studies.
What the trials show. The supplementation story is more complicated than the observational story. Large RCTs like VITAL (n=25,871) did not find that 2,000 IU/day reduced cancer or cardiovascular events in unselected adults. The trials that do show benefit tend to enroll vitamin D-low participants. The NIH ODS fact sheet summarizes the deficiency-correction case well: if your 25-OH D is under 30 ng/mL, supplementation is reasonable; if it is already 40 to 60, adding more is unlikely to do much.
Dose used in trials. 1,000 to 4,000 IU/day depending on baseline status. K2 (MK-7 form, 90 to 180 mcg) is commonly co-formulated; the human evidence for K2 is mechanistic plus modest bone-density and arterial-stiffness signal, not yet conclusive.
Form to look for. D3 (cholecalciferol) in an oil base for absorption. K2 as MK-7 if you want the combination.
Skip if your blood level is already above 30 ng/mL and you have no risk factors for low vitamin D.
Actionable takeaway: Get a serum 25-OH vitamin D test once before starting. Supplement to a target of roughly 40 to 60 ng/mL, not to infinity.
Magnesium (glycinate or L-threonate)
Why it helps. Magnesium is a cofactor for over 300 enzymes, including ATP-dependent reactions, and modulates NMDA receptor activity and calcium channel function in neurons. Magnesium L-threonate is a chelated form that crosses the blood-brain barrier more efficiently than oxide or citrate in animal models, which is the basis for its specific cognitive marketing.
What the trials show. The general magnesium signal in sleep and stress is reasonable: a 2022 systematic review found small to moderate effects on subjective sleep quality. The L-threonate-specific human data is thinner. A 2016 small RCT in older adults with cognitive complaints using magnesium L-threonate showed improvements on executive function and working memory, but the trial was small (n=44) and has not been independently replicated. Most of the supportive L-threonate data is animal.
Dose used in trials. 200 to 400 mg of elemental magnesium for sleep and stress (glycinate is the standard form); 1.5 to 2 g/day of magnesium L-threonate for the cognitive endpoint.
Form to look for. Magnesium glycinate for sleep, threonate if you specifically want the brain-targeted formulation. Magnesium oxide is poorly absorbed and not what trials used.
Skip if you have advanced kidney disease or are on a magnesium-containing antacid.
Moderate evidence: the parts with reasonable mechanism but smaller human trials
Ashwagandha (KSM-66)
KSM-66 is a standardized root extract with the largest RCT base of any ashwagandha preparation. The Salve et al. 2019 RCT and the Lopresti et al. 2019 RCT both used 600 mg/day for 60 days in stressed adults and showed reductions in perceived stress and morning cortisol. Mechanistically, ashwagandha appears to modulate GABA-A signaling and HPA-axis output. Skip if you have hyperthyroidism (it raises T4 modestly), are pregnant, or are on lithium.
L-theanine
L-theanine is an amino acid from green tea that crosses the blood-brain barrier and increases alpha-wave activity on EEG. The Hidese et al. 2019 RCT used 200 mg/day for four weeks in healthy adults and reported improvements on subjective stress and sleep quality. The mechanism touches glutamatergic and dopaminergic signaling in prefrontal cortex. Trial-style dose: 100 to 200 mg as needed. Combines well with caffeine for sustained attention.
Rhodiola rosea
A 2011 systematic review found Rhodiola reduced subjective fatigue in 6 of 7 small trials, but heterogeneity was high and blinding was variable. Rosavin and salidroside (the standardized actives) appear to modulate monoamine signaling and HPA-axis output in animal models. Trial dose: 200 to 600 mg/day of SHR-5 or equivalent extract. Worth a 4 to 8-week trial in adults with stress-related fatigue; do not expect a stimulant-sized effect.
Alpha-GPC
Alpha-GPC is a choline donor that increases acetylcholine availability. The Parker et al. 2010 trial used 600 to 1,200 mg/day and showed a peak power and growth hormone signal in young athletes. The cognitive-aging trials are older and less convincing for healthy adults. Skip if you have a personal or family history of stroke; observational data has flagged a possible association that is not settled.
Popular but evidence-thin
Tongkat Ali plus Fadogia agrestis
The "T-boosting stack" is heavily discussed in long-form podcasts. Tongkat Ali has some human RCT data, including a Singh 2022 trial in stressed adults showing modest testosterone increases at 200 mg/day over 12 weeks. Fadogia agrestis has essentially no published human trial data; the testosterone-raising claims rest on rat studies that also showed testicular toxicity at high doses. Tongkat Ali alone is a low-risk short trial in a stressed adult with low-normal testosterone. Fadogia at supplement-aisle doses is data-free, and the rat toxicity is a real reason to pause. Men with documented hypogonadism should see an endocrinologist before either.
Apigenin
Apigenin is a flavonoid found in chamomile, parsley, and celery, recommended at 50 mg pre-sleep. The human RCT base for apigenin as an isolated sleep aid is essentially nil. The mechanistic case (GABA-A modulation, anxiolytic signal in animal models) is real but small. Chamomile tea has a small RCT signal for sleep in older adults; the isolated apigenin pill is a confident-sounding extrapolation. Low-risk if you want to try it; do not expect a clear sleep effect.
Urolithin A (Mitopure)
Urolithin A is a metabolite of ellagic acid that promotes mitophagy. The published trial in middle-aged adults showed biomarker improvements at 500 mg/day Mitopure over four months, with smaller signals on muscle endurance. That is a surrogate-endpoint trial, not a clinical-outcome trial. Mechanistically interesting; the evidence is mostly biomarker and the cost is high.
What to look for when buying
| Question | What to check |
|---|---|
| Does the label show milligrams of the specific active, not just the herb name? | EPA mg, elemental magnesium mg, standardized extract percent and active mg |
| Third-party tested? | ConsumerLab approved, USP Verified, NSF Certified for Sport, Informed Sport |
| Form (for the specific compound)? | Creatine monohydrate (ideally Creapure), magnesium glycinate or threonate, KSM-66 ashwagandha, EPA-dominant TG-form fish oil, D3 in oil base |
| Is there a proprietary blend hiding the dose? | If yes, skip; you cannot match the trial dose without per-ingredient mg |
| Is the brand recommendation independent of a commercial partnership? | The Huberman/Momentous arrangement is disclosed but not independent verification |
For our framework on how we vet brands, see how we review supplements.
When supplements are not enough
Stop building a stack and book a clinician evaluation if:
- Insomnia is severe enough to impair daytime function for more than a few weeks (sleep medicine is a real specialty)
- Mood is low for two or more weeks with loss of interest, sleep change, or any thoughts of self-harm (in the US, call or text 988 for the Suicide and Crisis Lifeline)
- Energy and focus problems appeared abruptly in adulthood (rule out thyroid, sleep apnea, anemia, and depression before guessing at apigenin)
- Testosterone-related symptoms (low libido, fatigue, mood) are persistent, in which case an endocrinology workup precedes any Tongkat Ali experiment
- You are pregnant, nursing, or planning to be, as the safety profile of most of this stack in pregnancy is either uncertain or contraindicated
The real question is not "what is on the podcast," it is "what does my data say I am actually missing."
FAQ
Does Andrew Huberman take all of these every day?
He has described the full stack across multiple episodes, but on-mic he distinguishes between a daily core (creatine, omega-3, vitamin D, magnesium) and intermittent or cycled items (Tongkat Ali, Fadogia, Rhodiola). Most of the strongest-evidence supplements sit in his daily list.
Is Momentous the best brand for these?
Momentous is his disclosed brand partner, and several of their products meet third-party testing standards. That does not make them uniquely the best; Thorne, Klean Athlete, Nordic Naturals, and others meet the same criteria, often at lower cost.
How long before I should expect a noticeable effect?
Creatine: two to four weeks for a strength difference. Omega-3: 8 to 12 weeks for mood and cognition. Vitamin D: a serum-level retest at 8 to 12 weeks. Ashwagandha and Rhodiola: four to eight weeks.
Can supplements replace prescription medication for sleep, mood, or testosterone?
No. For clinical insomnia disorder, major depression, or hypogonadism, the standard of care is clinician evaluation and, where indicated, prescription treatment. Supplements are adjunctive.
Is Fadogia agrestis safe?
Human safety data is essentially absent, and rat studies have flagged testicular toxicity at the higher end of the dose range. Until human trials clarify the picture, we do not recommend it.
Conclusion: the bottom line on best supplements Andrew Huberman recommends
The honest summary: creatine monohydrate, EPA-dominant omega-3, vitamin D3 to a tested target, and magnesium for sleep are the four parts of the Huberman stack with the strongest human RCT support, and they happen to be among the cheapest items on it. Ashwagandha (KSM-66) and L-theanine have reasonable mid-tier data for stress and sleep. Alpha-GPC, Rhodiola, Tongkat Ali, apigenin, Fadogia, and urolithin A range from mechanism-plus-small-trial to essentially data-free in humans. The stack is not wrong; it is bigger than the evidence requires.
Next steps:
- If you take nothing else from the list, start with creatine monohydrate at 5 g/day and an EPA-dominant fish oil dosed to roughly 1,000 mg EPA
- Get a serum 25-OH vitamin D and a magnesium-status proxy (red cell magnesium or dietary intake review) before deciding on D3 and magnesium doses
- For an alternative longevity-framed stack to compare against, read our breakdown of the Peter Attia supplement stack, and see our author page for related cognitive and mood coverage
Reviewed by Maria Rodriguez, MS Nutrition Science, focused on cognitive and mood biochemistry.
This article is for informational purposes and not medical advice. Supplements can interact with prescription medications, including anticoagulants, SSRIs, lithium, and thyroid medications, and dosing recommendations are not one-size-fits-all. Consult a licensed physician before starting any supplement protocol, particularly during pregnancy or nursing, alongside prescription medications, or if you have a chronic health condition.
Recommended Products
As an Amazon Associate, UsefulVitamins.com earns from qualifying purchases at no extra cost to you. We only recommend products supported by published research or third-party testing.
