The State of Nootropics 2026

Chapter 1

The state of the field in 2026

Three things have changed in the last twelve months that matter for anyone taking cognitive enhancers seriously.

The DSHEA-era assumptions are getting reviewed. The FDA has signaled, repeatedly through 2025, that it intends to revisit the dietary supplement classification for compounds that look more like drugs than like food, particularly synthetic racetams, novel peptides, and high-bioavailability nicotine formulations. The 2018 phenibut and 2019 vinpocetine actions were the opening rounds; the 2024–2025 enforcement against tianeptine vendors was the second; the 2026 enforcement priorities now circulating in industry trade communications suggest the third round is coming. If you build personal protocols around supplement-channel availability of compounds the FDA could pull at any time, build them with that timeline in mind.

Modafinil has lost its uniqueness. Five years ago modafinil was the one drug everyone outside the ADHD-prescription population reached for. Today the picture is more crowded: armodafinil is meaningfully cheaper and arguably better-tolerated; flmodafinil sits in the supplement grey-market; and a handful of new Russian-origin compounds (bromantane, semax in spray form) are now reliably available through international vendors. The competitive pressure has compressed prices for all of them. The clinical evidence story has not changed materially, modafinil is still the most-studied and best-understood of the wakefulness-promoting agents, but the buying decision is more complicated than it was.

The longevity-cognition merge is now the dominant narrative. Anyone who has been around the space since the original "smart drugs" framing of 2010-2012 has noticed the framing shift. The dominant story in 2026 is not "I want to focus harder for this exam", it is "I want my brain to work in twenty years." The compounds being discussed reflect that: NAD+ precursors (NR, NMN), senolytics (fisetin, quercetin), mTOR modulators (rapamycin), and the autophagy-supportive interventions (spermidine, urolithin A). These overlap heavily with the cognitive-enhancement stack, but the framing, and the willingness to pay, is structural rather than acute.

Underneath these three changes, the substance-level evidence has continued to accumulate. As of June 2026, the high-quality clinical literature on at least the foundation compounds, caffeine, L-theanine, omega-3, vitamin D, magnesium, creatine, ashwagandha, is essentially settled. The questions worth asking about those substances are now individual ones (genotype, baseline, interaction with your specific situation) rather than population-level ones.

What is not settled, and where most of the genuinely interesting research is happening: the peptide class (BPC-157, semax, selank), the broader senolytic and autophagy space, and the question of whether any of this generalizes to long-term outcomes rather than acute biomarker change. We expect the next three years to produce more answers in these three areas than the last ten years combined.

---

Chapter 2

The substances people actually take

We catalog 160 substances on Nootropics.com. The vast majority of users will, in practice, use a small handful. Which handful?

In the absence of population survey data, we have not run one, we draw on three signals: the substances that account for the largest share of the citation and discussion record in the published literature; the substances that show up most frequently in our editorial stack library; and the substances most-listed by the vendor directory.

By that combined ranking, the core ten substances of 2026 are:

1. Caffeine, still the dominant cognitive enhancer by volume of use and by clinical literature; effect size on vigilance is among the largest in the entire space. 2. L-Theanine, the canonical caffeine pairing; the 1:2 caffeine:theanine ratio is, after fifteen years, still the most-validated cognitive stack in the published RCT literature. 3. Omega-3 (EPA/DHA), structural support for membrane fluidity and synaptic function; the chronic-use case for anyone whose dietary intake is below the modern Western average. 4. Vitamin D3, deficiency is the modal baseline state in temperate-latitude populations; the cognitive-mood-immune effects are downstream of correction rather than enhancement. 5. Magnesium (L-threonate or glycinate), sleep support and NMDA modulation; threonate has the higher central nervous system bioavailability; glycinate is cheaper and works for everyone who is not specifically chasing CNS-target effects. 6. Creatine, almost-forgotten in the cognition discussion until the 2021-2024 wave of evidence on cognitive effect at standard 5g daily; the cheapest and best-validated supplement on this list. 7. Bacopa monnieri, the most-evidenced botanical for memory; takes 8-12 weeks to show effect; the gold standard against which newer memory compounds are measured. 8. Lion's Mane (Hericium erinaceus), NGF and BDNF support; the human evidence is thinner than the marketing suggests but the safety profile is strong and the mechanism is real. 9. Ashwagandha (KSM-66 standardization), the most-validated adaptogen; effect on stress, sleep, and downstream cognition is robust across multiple well-conducted trials. 10. Modafinil, for the prescription-eligible population, the most-effective single agent for wakefulness; cycling and individual tolerance still matter.

What is striking about this list is how much of it would have been on the equivalent list in 2020, in 2016, or in 2012. The center of gravity has not moved as much as the noise around the edges suggests. The newer, flashier compounds, Qualia, Mind Lab Pro, the various proprietary blends, are commercial repackagings of these foundations with niche additions.

The list of what people are talking about and what people are actually buying have diverged sharply. The discussion-volume leaders of 2026 are NR/NMN, the senolytics, peptides, and the longevity stack, but the purchase volume leaders are the substances above, with caffeine alone outweighing every novel compound combined.

---

Chapter 3

Stack patterns

The dominant stack architectures of 2026 cluster into four patterns. Anyone designing a personal protocol is, whether they know it or not, picking from this list.

Pattern 1: The foundation stack. Three to five substances, all foundational, all with A-grade evidence, all taken daily without cycling. Caffeine + L-theanine + magnesium + omega-3 + vitamin D is the canonical version. Cost: under thirty dollars a month. Effect: small on each axis individually, additive across the bundle, sustainable indefinitely. This is, by any reasonable measure, the highest expected-value choice for users who are not specifically chasing a particular cognitive symptom.

Pattern 2: The cognitive acute stack. Built around a wakefulness agent (caffeine, modafinil, or armodafinil) plus a calming complement (L-theanine, possibly L-tyrosine if depletion is a concern) plus a cholinergic (Alpha-GPC or CDP-choline). Used on work days, not on rest days. Cycling is built into the protocol. This is the stack for users with clear performance windows, exams, deadlines, focus work, where the marginal value of acute cognitive effect is high.

Pattern 3: The chronic-build stack. Bacopa + Lion's Mane + Ashwagandha, taken daily for 8-12 weeks, evaluated for effect at the end of that window. The mechanism is structural, these compounds modulate gene expression, dendritic branching, BDNF and NGF, and the effect is not visible day-to-day. The discipline cost is real: most users who say a chronic compound "did nothing" stopped at four weeks. The actual evidence base on chronic compounds is much stronger than the dropout-rate-skewed user reports would suggest.

Pattern 4: The longevity-cognition merge stack. NR or NMN + a senolytic protocol (fisetin pulse, quercetin daily) + the longevity-specific antioxidants (CoQ10, PQQ) + low-dose rapamycin under medical supervision. Optimized not for acute cognitive effect but for trajectory over decades. The evidence here is much earlier-stage than the foundation literature; the willingness to pay is much higher; and the user demographic is older and more comfortable with the risk-of-novelty trade.

Most experienced users converge on a combination of Pattern 1 (foundation, daily, no thinking required) plus situational layering from Pattern 2 (on demand) or Pattern 3 (chronic projects) depending on goals. Pattern 4 is the recent addition and is increasingly bolted on for users in the 40+ demographic.

The most common stacking mistake we still see: adding three substances at once, evaluating after a week, and either over-attributing effect or stopping the whole thing because something didn't work. The disciplined add-one-substance-at-a-time-for-five-days protocol remains the highest-information design.

---

Chapter 4

Sleep and cognition, the hidden multiplier

If we had to identify the single highest-leverage intervention in the entire cognitive enhancement space, it would not be a nootropic. It would be sleep.

The evidence is unambiguous and has been for a decade. A single night of restricted sleep impairs cognitive performance more than no nootropic improves it. Chronic sleep restriction at six hours per night for a fortnight produces cognitive deficits indistinguishable from acute total deprivation. The substances we discuss in this report cannot meaningfully outperform what fixing sleep does.

This is not a new insight. What has changed in 2026 is that the supplement-channel sleep interventions have improved enough to be a meaningful piece of the puzzle.

The current sleep stack, in roughly the order in which we would add things:

- Magnesium glycinate or L-threonate, 200-400 mg, 30-60 minutes before bed. Magnesium status is widely deficient in modern populations; correcting it has measurable effects on sleep latency and depth. - Apigenin, 50 mg, before bed. A botanical-flavonoid GABA-A modulator with mild anxiolytic effect at sleep onset. The David Sinclair recommendation has driven a wave of awareness; the underlying evidence is modest but the safety profile is excellent. - Glycine, 3 g, before bed. Lowers core body temperature, which is one of the levers sleep onset depends on. Cheap, well-tolerated, well-studied. - L-theanine, 100-200 mg, before bed. The same substance that pairs with caffeine works on the other end of the day, anxiolytic without sedation, useful for users whose sleep is being delayed by rumination rather than by physical state.

Notably absent from this list: melatonin. Melatonin has well-documented efficacy as a phase-shift agent for jet lag and shift work, but it is widely misused as a general sleep aid. At the doses sold in the US (often 5-10 mg) it produces measurable downstream effects on hormonal rhythms that are not benign over chronic use. If you are using melatonin nightly, the European 0.3 mg dose is closer to physiological replacement than the US 5 mg is.

The single highest-impact intervention in this category is none of the above: it is light. Bright light exposure within 30 minutes of waking, low-light or amber-light exposure in the three hours before sleep. The cost is zero. The effect on sleep architecture exceeds anything in the supplement category. If you are not already doing this, no nootropic stack will make up the deficit.

---

Chapter 5

The longevity-cognition convergence

Five years ago the cognitive enhancement community and the longevity community were two largely separate populations with two largely separate compound lists. In 2026 they have substantially merged.

The mechanism of the merge is biological. The pathways involved in cognitive aging, mitochondrial decline, accumulation of senescent cells, autophagy dysfunction, NAD+ depletion, are the same pathways involved in systemic aging. A compound that addresses any of them will, plausibly, have effects across both domains. The cognitive enhancement community has, accordingly, picked up the longevity stack; the longevity community has picked up the foundation cognitive stack.

The 2026 longevity-cognition stack converges on four mechanisms:

1. NAD+ restoration. NR (nicotinamide riboside) and NMN (nicotinamide mononucleotide) are the two precursors driving this category. NMN's history with the FDA, declared not a legal dietary supplement in late 2022, with enforcement uneven since, has pushed the market toward NR for US consumers and toward NMN for international consumers. The clinical evidence on cognitive effects of NAD+ restoration is suggestive but not conclusive; the mechanistic story is strong.

2. Senescent cell clearance. Fisetin is the senolytic of choice in the consumer market, pulse dosing (high doses 2 days per month) is the protocol the published animal-model work points to. Quercetin and dasatinib appear in the clinical research literature but dasatinib is a prescription kinase inhibitor and is not a supplement-channel option. The cognitive effects of senolytics in humans are not yet established; the systemic effects on aging biomarkers are, in some studies, striking.

3. mTOR modulation. Low-dose intermittent rapamycin is the central intervention. This is a prescription drug and requires a clinician familiar with its longevity-context use; it is not a supplement-channel option. The cognitive effects are not the primary motivation, the systemic anti-aging case is, but the cognitive downstream effects of healthy mTOR balance are real.

4. Autophagy support. Spermidine, urolithin A, and the broader polyamine pathway. Spermidine is the most-evidenced and most-accessible; urolithin A is a more recent entrant with a strong commercial push and an evidence base that is growing fast. The mechanism is well-defined (autophagy upregulation); the cognitive and systemic effects in humans are still being characterized.

The honest read in 2026: this category is two to five years away from having the kind of clinical evidence base that the foundation stack has. The mechanistic case is solid. The trajectory of the evidence is positive. The risk of one or more of these compounds turning out to be net-zero (or net-negative) over the long-term is real and worth pricing in.

Users in the 40+ demographic who can afford to be wrong about one of these compounds are the right population to be experimenting. Younger users without specific indications would, in our editorial read, get more out of getting the foundation stack right than from chasing the longevity-cognition merge.

---

Chapter 6

The grey market

The grey market did not go away. It got bigger.

Modafinil sourced from international pharmacies, peptides from research-chemical vendors, racetams and noopept from European mail-order pharmacies, phenibut from any of the half-dozen vendors that have replaced the ones that got shut down, these are not edge-case behaviors. Some meaningful fraction of the user population we serve is making at least one of these purchases at least once a year.

We do not facilitate these purchases on Nootropics.com. We track them, we cover the legal status, we publish the safety profile, and we refuse to link to vendors of controlled substances even when we are asked.

The honest assessment of the grey market in 2026:

Modafinil from international pharmacies continues to be the most-purchased grey-market item. Quality (where verified) has been broadly acceptable for a decade; the legal risk is real but enforcement against individual import has been minimal. We see no evidence that the FDA or DEA are prioritizing personal-import-level enforcement; we also see no evidence that the situation is going to remain stable indefinitely. Plan accordingly.

Peptides from research-chemical vendors have grown substantially in the past two years. BPC-157, TB-500, semax, selank, and the various combinations of these are the volume drivers. Quality is highly variable; third-party testing infrastructure is minimal; mislabeling is common. The clinical evidence in humans is thin enough that the risk-reward calculation is genuinely uncertain. We recommend that anyone in this category demand certificate of analysis with specific lot numbers and test independently where the dose is high enough to matter.

Phenibut remains widely available and continues to produce a steady stream of emergency department presentations for withdrawal, dependence, and acute overdose. The compound is not safe at the doses typical users land on after even a few months. We have not changed our editorial position: this is the one substance in the broader nootropic category we explicitly recommend against, regardless of the user's risk tolerance.

Kratom sits in a separate category. Legal in most US jurisdictions, banned in some, restricted in others. The substance is genuinely useful at low doses for stimulant-anxiolytic-pain effects; the risk of dependence at higher doses is substantial and is materially under-disclosed in the consumer market. The 7-hydroxymitragynine-extract products that have appeared in 2025-2026 are a meaningfully more dangerous category than the raw leaf material and should be approached with the caution due to a partial mu-opioid agonist.

The grey market exists because the regulatory infrastructure has not kept up with what the consumer market actually wants. We do not expect it to shrink in 2026. The most useful editorial intervention is honest reporting of risk profiles; the second most useful is the framing that whatever you import is your responsibility, full stop, and the FAQ that vendors will not refund seized parcels.

---

Chapter 7

AI advisor patterns, what people ask

Nootropics.com runs an AI advisor, a Claude-powered conversational interface against the substance and stack database. We launched it in late 2025. We do not log individual conversations in a way that allows us to share specific prompts. We do have visibility into the broad categories of question being asked.

The dominant question patterns of 2026:

"What should I take for [X]?" is the modal first message, where X is some combination of focus, energy, sleep, anxiety, memory, motivation, or a specific situation (studying, shift work, post-COVID). The honest answer in the majority of cases is some version of "tighten the foundation first", sleep, light, exercise, baseline nutrition, and the AI advisor has been trained to give that answer rather than to start naming compounds.

"Is X safe with my prescription medication?" is the second most-common pattern and the one where the AI advisor is genuinely useful, since the interaction matrices are exactly the kind of structured cross-referencing computers handle well. The interaction-checker tool on the site is more authoritative than the AI advisor on this; we route advisor conversations toward it.

"What's the difference between X and Y?" is the third pattern. The comparison-page generation on the site (modafinil-vs-armodafinil, NR-vs-NMN, bacopa-vs-lions-mane) was driven in part by recognition of how dominant this pattern was in early advisor conversations.

What we have learned, watching the advisor in production for six months: most users do not want a recommendation. They want a framework for making their own decision. The advisor is most useful when it surfaces the relevant data, half-life, evidence grade, interaction risk, expected onset, rather than when it produces a confident prescription.

This is an editorial commitment that has gotten clearer through 2026: we are not in the business of telling users what to take. We are in the business of producing the substance-level information dense enough that users can make their own decisions, and giving them tools that let them check their own work.

---

Chapter 8

What we got wrong in 2025

The previous year of editorial work on Nootropics.com produced several positions we have updated.

On NMN. Our 2024 and early 2025 coverage of NMN treated it as a high-promise compound with a clear regulatory path. The FDA's late-2022 reclassification of NMN as not-a-legal-dietary-supplement was less significant in our coverage than it should have been, and our framing of NR-versus-NMN under-disclosed the enforcement risk specifically attached to the NMN side of the comparison. We updated the comparison page in February 2026 to reflect this. The substance itself remains mechanistically interesting; the regulatory layer is more material than we previously communicated.

On the racetam-prescription situation in Europe. Our coverage of racetams (piracetam, aniracetam, oxiracetam, pramiracetam) framed them as "unscheduled in the US" without sufficient emphasis on the prescription-only status across most of the EU and Australia. This produced situations where international users, reading our default US-framed coverage, ordered compounds that were intercepted at customs. The country-specific buying guides we shipped in mid-2026 are the structural fix to this.

On the rate at which the longevity-cognition merge would happen. We thought, in late 2024, that the cognitive and longevity communities would remain meaningfully distinct through the decade. The merge happened faster than we expected. The substance-level analysis is the same in both directions, but the audience-level framing, who is buying these compounds and why, turned out to be more unified than we predicted.

On Lion's Mane. Our coverage was, for several years, more confident in the human cognitive effects of Lion's Mane than the clinical evidence base supports. The animal-model NGF and BDNF data are genuinely strong. The human-cognitive-effect data is thinner than the popular narrative implies. We updated the Lion's Mane entry to reflect the actual state of the literature in mid-2025; the editorial position is now appropriately cautious about acute effects while still positive on the long-term safety profile.

We expect to update this section every year. The editorial honest discipline is naming, in the public record, when our earlier positions were wrong, and explaining what changed.

---