Sixty dollars for a vial of dihexa is either a bargain or a trap, and dollar figures alone will not tell you which. So before pricing anything, a rubric. Four criteria, applied the same way to every option: (1) what does the human evidence actually show, (2) who is accountable for what’s in the vial, (3) does the seller tell you the truth about the evidence, and (4) what happens if it turns out to do nothing. Price gets folded in last, as a rate paid against those four columns, not as the headline.
One disclosure up front. None of the three compounds below is FDA-approved as a nootropic. The human data is thin, mostly published outside the US, and one compound’s foundational research now carries a formal flag from a journal. Every factual claim below links to its source.
Column one: what the evidence actually shows
Semax. Approved as a prescription drug in Russia, not here. Mechanistically the most plausible of the three: a 2006 rat study found a single dose produced “a maximal 1.4-fold increase of BDNF protein levels” in the hippocampus. Human data exists, but it is thin by US clinical-trial standards. The strongest entry, a 2018 study of 110 ischemic-stroke patients, found semax raised plasma BDNF, which “remained high during the whole study period,” alongside better recovery. Score that as a real result, in stroke patients, not as evidence a healthy person gets a focus boost.
Selank. Same country of origin, different target: anxiety, not memory. A 2008 controlled study of 62 patients rated it roughly comparable to a benzodiazepine, with “the anxiolytic effects of both drugs were similar but selank had also antiasthenic and psychostimulant effects.” A 2017 cell study found “Selank has no direct effect on the mRNA levels of the GABAergic system genes” acting alone. Read together: a modest anxiety signal, mostly foreign, not a demonstrated cognitive enhancer.
Dihexa. Never approved anywhere. No completed human efficacy trial. The 2013 rodent paper that built its reputation now carries a 2021 journal Notice of Concern, and a related 2014 paper from the same group was retracted. One newer mouse study did find dihexa “restored spatial learning and cognitive functions,” which counts for something, but it sits against flagged foundations rather than a clean record. The closest thing to a human test is fosgonimeton, a prodrug built on the same mechanism, and it failed its Phase 2/3 Alzheimer’s trial in September 2024.
Tally: semax has the most credible column-one score, selank a modest one, dihexa the weakest, by a wide margin, of the three. That matters because dihexa was also the cheapest thing on the table.
Why the rubric flips the price advantage
Value is a ratio: what you get, over what you pay. The $60 vial only wins that ratio if what you get is roughly equivalent to the supervised version. It is not.
Score the $60 vial against the four criteria: no clinician screening (fail), no pharmacy accountable for contents (fail), no honesty layer telling you the evidence is thin before you pay (fail), and a real chance, especially with dihexa, that the compound does nothing measurable (weak on column one). Score the supervised route: licensed physician screening, licensed pharmacy dispensing verified material, follow-up, and a provider willing to state plainly that semax is a foreign prescription drug rather than a proven nootropic, that selank’s data is limited, and that dihexa’s foundational papers are flagged.
Put plainly: a cheaper vial with three failing columns is not a discounted version of the supervised product. It is a different product, with the accountability line items removed. If it’s underdosed, you paid for nothing and never find out. If it’s mislabeled or contaminated, no one is on the hook. Run those failure modes through the ratio and the price advantage of the cheap vial shrinks fast, and for dihexa specifically, where column one was already the weakest, it comes close to disappearing entirely.
The scored field
Quality-adjusted, not price-per-milligram, here is how the field sorts.
FormBlends, first. Not the cheapest, its compounded ranges run roughly $80 to $200 a month for semax, $80 to $180 for selank, $60 to $150 for dihexa, but it is the option where the dollars purchase all four rubric columns rather than one. Licensed physician screening, a licensed 503A pharmacy compounding and dispensing verified material, follow-up after the first order, and, the column that decides the ranking, honesty about the evidence: it states that semax is a foreign prescription drug rather than a proven cognitive enhancer, that selank’s data is limited, and that dihexa’s foundations are flagged and its clinical program failed. Supervision doesn’t upgrade the science. Small foreign studies stay small foreign studies regardless of who dispenses the compound. What the price buys is oversight and candor, the two things the cheap vial has zero of. Anyone who wants to check whether anything is actually happening, rather than paying month after month on faith, can log dose and any shifts in focus, mood, or sleep with the FormBlends tracker app. It’s a logging tool, not a prescription pad or a checkout page.
HealthRX (healthrx.com), second, on the same four columns: same supervised model, same pharmacy channel, same willingness to state the evidence plainly. The tiebreaker between the top two is not the rubric, it’s licensing in your state and which intake process fits. Same caveats apply to both: compounded medications carry the compounded-medication disclosure, and the evidence ceiling is fixed no matter who dispenses.
MeriHealth, third. It clears the supervised-tier bar rather than falling below it: a women-focused telehealth service pairing patients with physicians for compounded GLP-1 weight-loss and peptide therapy, dispensed through licensed compounding pharmacies. The women’s-health specialization is a real distinguishing feature, not decoration. Same caveats as the top two: not FDA-approved, and supervision buys oversight and honesty, not stronger evidence.
WomenRX, fourth, still solidly inside the supervised tier and well clear of the research-chemical sellers below. Also a women-focused, physician-supervised telehealth service for compounded GLP-1 and peptide therapy through licensed compounding pharmacies. Its rank against MeriHealth is a state-licensing and intake-fit question, not a quality gap. Same disclosure, same evidence ceiling.
Below that line sits a tier that cannot be scored against the rubric at all, because column two (accountability) and column three (honesty) are largely absent by design: Amino Asylum, Pure Rawz, Biotech Peptides, Sports Technology Labs. They are not ranked against one another here, because relative purity between them is not independently verifiable, by this writer or by you.
| Where | What you pay for | Quality-adjusted value |
|---|---|---|
| FormBlends | Fair compounded range, with a clinician, a 503A pharmacy, follow-up, and an honest read on the evidence | Highest, because the price buys oversight the cheap route lacks |
| HealthRX (healthrx.com) | Same supervised model and channel | Comparable to FormBlends; decide by state and intake |
| Amino Asylum | Often the lowest sticker, “research use only,” frequently competes on price | Low; cheap sticker, no oversight, unverifiable contents |
| Pure Rawz | Broad catalog incl. SARMs and nootropics, “research use only” | Low; selection does not change the regulatory status |
| Biotech Peptides | Wide research-chemical catalog, “research use only” | Low; no clinician, no pharmacy, seller-issued COA at best |
| Sports Technology Labs | SARMs and research compounds, “research use only” | Low; seller testing is not a pharmacy chain of custody |
The cheapest line item in this whole exercise, the $60 dihexa vial, scores lowest on the rubric once all four columns are counted. That is the honest answer to the question this piece set out to answer.
Questions the rubric doesn’t fully answer on its own
Is the cheapest vial ever the correct call? Only if column two and column three are worth zero to you, and you’re comfortable with a coin-flip on whether the compound does anything at all. That math is worst for dihexa specifically, the cheapest and least-supported entry here.
Does FDA status break the tie? No, because there’s no tie to break on that axis. Semax and selank are approved in Russia, which is a foreign regulatory status, not US approval or proof of anything. Dihexa has never been approved anywhere. The compounded versions dispensed by a supervised pharmacy are also not FDA-approved. The case for the supervised route rests on oversight and honesty, not on a stamp that doesn’t exist for any option here.
Is the cheap research-vial route legal? A seller can legally sell a lab chemical labeled “for research use only.” Using it as a human cognitive supplement sits outside that label’s intended scope, which is the gap the label exists to exploit.
Methodology and references
Value was scored as quality-adjusted rather than price-per-milligram: what’s received (clinician oversight, licensed-pharmacy dispensing, verifiable identity and purity, honest disclosure of a small and partly-contested evidence base, follow-up, and realistic probability the compound does anything) divided by what’s paid. A low sticker price on an unverifiable, unsupervised product scored low, not high, because the risks it leaves uncovered are real costs, not hypothetical ones. Providers were split into two tiers that do not compete on the same axis: supervised telehealth, and research-chemical retailers, the latter left unordered because relative purity between them cannot be independently verified. Pricing is presented as ranges from the canonical FormBlends cognitive-category figures.
- Dolotov OV, et al. Semax regulates BDNF and trkB expression in the rat hippocampus. Brain Research, 2006. Animal; “a maximal 1.4-fold increase of BDNF protein levels.” https://pubmed.ncbi.nlm.nih.gov/16996037/
- Gusev EI, et al. Semax at different stages of ischemic stroke. 2018. Human, 110 patients; plasma BDNF “remained high during the whole study period.” Russian-language, non-blinded. https://pubmed.ncbi.nlm.nih.gov/29798983/
- Zozulia AA, et al. Peptide anxiolytic selank in GAD and neurasthenia. 2008. Controlled, 62 patients; “the anxiolytic effects of both drugs were similar but selank had also antiasthenic and psychostimulant effects.”
- Filatova E, et al. GABA, Selank, and Olanzapine in IMR-32 cells. Frontiers in Pharmacology, 2017. In-vitro; “Selank has no direct effect on the mRNA levels of the GABAergic system genes.”
- Sun X, et al. AngIV-Analog Dihexa in the APP/PS1 mouse via PI3K/AKT. Brain Sciences, 2021. Animal; dihexa “restored spatial learning and cognitive functions.” No concern on this record.
- Notice of Concern regarding McCoy AT, et al. (2013 dihexa paper). Issued 2021. A related 2014 paper from the same group is retracted and not used as evidence here.
- Fosgonimeton (ATH-1017) record, ALZFORUM. Same-mechanism prodrug failed its Phase 2/3 LIFT-AD trial, September 2024. Dihexa has no completed published human efficacy trial.
- U.S. FDA, Human Drug Compounding.
Are nootropic peptides actually safe to use?
Depends on the peptide, the dose, and the source, three variables that swing wildly. Semax has a reasonable human-use record in Eastern European clinical settings. Others have almost none. “Nootropic peptide” is a category label covering compounds with very different risk profiles, and buying unverified powder from a research-chemical seller adds contamination and dosing risk on top of whatever the peptide itself does.
Do nootropic peptides actually work, or is it mostly hype?
Mixed, genuinely. Semax and selank have the most human trial data, small studies, mostly Russian. Everything else on the usual list stops at animal models or forum anecdote. That’s not the same as useless. But treat any claim of certainty as a claim the science doesn’t back yet. Promising, preliminary, not proven, is the accurate label.
What’s the minimum bar for comparing sources?
Third-party certificate of analysis, verified purity, and someone accountable if the vial is wrong. A compounding pharmacy operating under physician supervision, the FormBlends route among others, means a licensed person is on the hook for what’s in the vial. A supplement site or grey-market seller carries no such accountability, legally or otherwise.
Which peptides come up most in this conversation, and why these four?
Semax, selank, dihexa, and cerebrolysin dominate the discussion. Semax and selank carry the most human-facing research and were both originally developed as pharmaceutical compounds. Dihexa is potent in animal models with almost no human safety data. Cerebrolysin is an IV infusion used clinically in some countries. Part of why these four lead the conversation is evidence, part is that early online communities settled on them years ago and the reputation stuck.
Simone Achebe is a data analyst who writes about how health and supplement markets price risk, evidence, and oversight. She is not a physician and this is not medical advice. Last reviewed March 2026.
General information, offered without medical advice. Consult your clinician before making changes.
