The best peptides for cognition, honestly graded, have little human evidence behind them. No peptide is FDA-approved or proven to improve memory, focus or thinking in healthy people. Semax and Selank have small Russian studies, mostly in stroke and anxiety rather than healthy adults.[2][8] Cerebrolysin has short positive trials in Alzheimer’s disease.[10] Dihexa and P21 have animal data only, with no human trials.[13][17] The best-supported way to protect cognition is managing known risk factors such as blood pressure, hearing loss and inactivity.[1]
This guide grades each compound. For more, see our nootropic peptides hub.
Evidence at a glance
| Peptide | Evidence level | What was shown | FDA status | Our profile |
|---|---|---|---|---|
| Risk-factor management (not a peptide) | Human RCTs and cohort data | 14 risk factors linked to about 45% of dementia[1] | n/a | n/a |
| Cerebrolysin | Small human studies (in dementia) | Short-term gains in small Alzheimer’s trials[10] | Not approved | Cerebrolysin |
| Semax | Small human studies | Brain-network changes on scans; no memory trial[4] | Not approved | Semax |
| Selank | Small human studies | “Mild nootropic effects” reported in anxiety patients[8] | Not approved | Selank |
| Dihexa | Animal or lab only | Better memory in rat models[13] | Not approved | Dihexa |
| P21 | Animal or lab only | Better learning in mice and aged rats[17][18] | Not approved | P21 |
Start with what is proven
There is no approved “smart drug” for healthy people. For lowering dementia risk, the strongest evidence points to everyday health. The 2024 Lancet Commission on dementia linked about 45% of cases to 14 modifiable factors, including less education, hearing loss, high LDL cholesterol, depression, head injury, physical inactivity, diabetes, smoking, high blood pressure, obesity, heavy drinking, social isolation, air pollution and untreated vision loss.[1]
If memory problems are already affecting daily life, a clinician can test for treatable causes and, where appropriate, discuss approved Alzheimer’s treatments. See peptides for Alzheimer’s and Parkinson’s.
Semax: the most-used “nootropic peptide”
Semax is a nasal peptide based on a fragment of the hormone ACTH. It is a prescription drug in Russia, where its label covers stroke and other brain conditions.[2]
- Animal data: in rats, Semax increased BDNF, a growth factor that supports brain cells and learning.[3]
- Human data: in 24 healthy volunteers, one nasal dose changed activity in the brain’s default mode network on MRI.[4] A separate study of 52 healthy people found Semax and Selank each changed connectivity between the amygdala and temporal cortex.[5] Neither study measured whether memory or focus improved.
FDA staff reviewed Semax in 2026 and recommended against allowing it in pharmacy compounding.[6] Advisers reportedly voted 8 to 5 in favor anyway, and FDA has not made a final decision.[7]
Selank: an anxiety peptide with nootropic claims
Selank is sold in Russia for anxiety. In a 60-patient Russian study, authors reported “pronounced anxiolytic and mild nootropic effects.”[8] In another, adding Selank to a benzodiazepine reduced that drug’s memory and attention side effects.[9] These were small studies in anxious patients, not cognitive-enhancement trials. See best peptides for anxiety.
Cerebrolysin: dementia data, not enhancement
Cerebrolysin is an injected mix of peptides from pig brain, used in Russia and parts of Europe and Asia.
- A 2015 meta-analysis of six small trials in mild to moderate Alzheimer’s found better cognition at 4 weeks and better global ratings at 6 months.[10]
- A 3-year Russian study of 100 people with mild cognitive impairment reported fewer conversions to dementia with yearly infusion courses, but it was open-label and the comparison group got no placebo.[11]
- In acute stroke trials, Cochrane found a possible increase in non-fatal serious adverse events.[12]
It requires repeated injections or IV infusions and has not been tested in healthy people.
Dihexa: big claims, no human data
Dihexa was designed from angiotensin IV and improved memory in rat models of dementia.[13] But:
- FDA says it has found no human exposure data for Dihexa by any route.[16]
- A 2014 paper describing how Dihexa works through the HGF/c-Met pathway was retracted in 2025.[14]
- A related drug acting on the same pathway, fosgonimeton, failed its primary and secondary goals in a 287-person Alzheimer’s trial.[15]
Read more in Dihexa vs Semax vs Selank.
P21: mouse and rat studies only
P21 is a small peptide derived from a nerve growth factor called CNTF. It improved learning and memory in normal mice and reduced age-related memory decline in old rats.[17][18] It has never been tested in people, and no human trial of P21 is registered.
Why “nootropic peptide” claims outrun the data
Most marketing for cognitive peptides borrows from three kinds of research: rat studies, brain scans in small groups of healthy volunteers, and trials in people with a disease such as stroke or dementia. Each has limits for someone hoping to sharpen memory or focus.
- Rat studies often fail to predict human results. Dihexa’s and P21’s strongest data are in rodents.[13][17]
- Brain-scan changes show a drug did something, not that thinking improved. The Semax and Selank imaging studies did not test memory or attention.[4][5]
- Disease trials don’t transfer to healthy brains. Cerebrolysin’s dementia trials measured decline in illness, not enhancement.[10]
A product that skips all three steps and goes straight to online sale has no evidence base at all.
Safety concerns
- Unknown contents: drug-control labs found Selank and Semax in seized products and noted they are sold online without completed trials.[19]
- Theoretical risks: Dihexa activates a growth pathway also linked to some cancers, and there are no human data to judge that risk.[16]
- Regulation: none is FDA-approved, and research peptides are not checked for purity or dose.
For related reading, see Selank vs Semax, peptides for ADHD and peptides for TBI and concussion.
