Semax is unusual among the compounds on this site in having a registered product, a clinical literature and a set of stated doses — and unusual in the way those doses are written. They are percentages and drops, because the registered form is a nasal solution rather than a vial of powder. That makes the arithmetic here a translation problem rather than a sourcing one: a percentage is a concentration, a concentration is what a reconstituted vial produces, and the two can be compared directly. This page reports what the trials and the registrations state, converts each figure into mg/mL and U-100 syringe units, and flags one published figure that cannot be read as printed. It reports and does not recommend.
The Semax calculator runs the sums for any vial and volume; the water volumes for the 10 mg and 30 mg vials are worked through on the Semax reconstitution page, and the all-compounds chart carries the one-line version beside every other compound.
The percentage is the concentration
A percentage solution in this notation is grams per 100 millilitres. So 0.1% is 0.1 g in 100 mL, which is 1 mg/mL, and 1% is 10 mg/mL. Both registered Semax strengths are therefore ordinary concentrations, and a vial of powder reaches either of them by division:
| Target | What it means | 10 mg vial | 30 mg vial |
|---|---|---|---|
| 0.1% | 1,000 mcg/mL | 10 mL of water | 30 mL |
| 0.5% | 5,000 mcg/mL | 2 mL | 6 mL |
| 1% | 10,000 mcg/mL | 1 mL | 3 mL |
The 10 mL and 30 mL rows are arithmetic rather than options: research vials are rarely built to hold that much liquid, and a volume the vial cannot take is not a volume. The practical consequence runs the other way — a vial reconstituted in the 1 to 3 mL that fits it is at or near the 1% strength, which is the strength the largest clinical figures were given at.
What the trials state
| Study | Subjects | Dose and route | What it reported | Opened |
|---|---|---|---|---|
| Zh Nevrol Psikhiatr, PMID 11517472 (1997) | 30 patients, acute hemispheric ischaemic stroke; 80 controls | 12 mg/day (moderate) and 18 mg/day (severe), intranasal, 5 and 10 day courses | Some influence on the rate of restoration of neurological function, chiefly motor | 2026-09-20 |
| Zh Vyssh Nerv Deiat, PMID 9173745 (1997) | Review of fifteen years of research, animals and people | 0.015 to 0.050 mg/kg intranasal | Effects on memory, attention and hypoxia resistance lasting 20 to 24 hours | 2026-09-20 |
| Vestn Oftalmol, PMID 10741256 (2000) | Patients with optic nerve disease, three groups | Nasal drops, or endonasal electrophoresis, added to standard therapy | Improved visual acuity and visual field against control | 2026-09-20 |
| Zh Nevrol Psikhiatr, PMID 18379501 (2007) | 27 patients, motor neuron disease | 1% solution, 12 mg/day, two 10-day courses, two-week break | No effect on denervation or clinical scores; quality-of-life total improved, maximal at day 10 | 2026-09-20 |
| Bull Exp Biol Med, PMID 30225715 (2018) | 24 healthy volunteers, mean age 43.9 | 1% solution intranasally, single administration | A larger rostral subcomponent of the default-mode network on resting-state fMRI against placebo | 2026-09-20 |
| Zh Nevrol Psikhiatr, PMID 29798983 (2018) | 110 patients after ischaemic stroke | 6,000 mcg/day, two 10-day courses, 20-day interval | Plasma BDNF, motor performance and Barthel index compared across early and late rehabilitation | 2026-09-20 |
Two features of that set matter more than any single number. The range is wide — 6 mg a day in rehabilitation, 12 to 18 mg in the acute period, and a per-kilogram figure from the research literature that comes to 1 to 3.5 mg for a 70 kg adult — and the spread is not noise but three different questions. And the strongest design in the group is an open-label study of 27 people; there is no large blinded trial here to anchor any of it.
The figure that cannot be read as printed
A 2015 Russian study of lipid metabolism in psoriasis reports that 60 patients received "0.1% semax solution intranasally 600 mg/day for 10 days". Those two figures contradict each other. At 0.1%, a millilitre holds 1 mg, so 600 mg is 600 mL of solution in a day — through the nose, for ten days. The reading consistent with the product named in the same clause is 600 micrograms a day, which is 0.6 mL of the 0.1% solution and sits sensibly between the rehabilitation and per-kilogram figures.
This is worth stating rather than quietly correcting, because it is the failure mode of the whole microgram-milligram family: the two units differ by a factor of a thousand, the symbols differ by one letter, and nothing in a citation chain re-checks the arithmetic. A figure copied from that sentence into a summary page arrives with a printed source and is wrong by three orders of magnitude.
The arithmetic per vial
The calculator's presets are the 10 mg and 30 mg vials. One U-100 unit is a hundredth of a millilitre (MedlinePlus, "Giving an insulin injection", reviewed 2024-07-21). The right-hand columns are a drop of the 0.1% product, the rehabilitation study's daily amount and the acute-stroke figure.
| Vial and water | mcg/mL | mcg per unit | 50 mcg | 6,000 mcg | 12,000 mcg |
|---|---|---|---|---|---|
| 10 mg in 1 mL (1%) | 10,000 | 100 | 0.5 units | 60 | 120 |
| 10 mg in 2 mL | 5,000 | 50 | 1 | 120 | 240 |
| 10 mg in 3 mL | 3,333 | 33.3 | 1.5 | 180 | 360 |
| 30 mg in 3 mL (1%) | 10,000 | 100 | 0.5 | 60 | 120 |
| 30 mg in 6 mL | 5,000 | 50 | 1 | 120 | 240 |
| 30 mg in 10 mL | 3,000 | 30 | 1.7 | 200 | 400 |
The right-hand column is the arresting one: at every volume that fits the vial, the acute-stroke daily amount is more than a full 1 mL insulin syringe, and at the 1% strength it is 1.2 mL — more than an entire 10 mg vial holds. The clinical figures were given as nasal drops of a manufactured solution, where 1.2 mL a day is unremarkable; expressed as syringe draws from a research vial, the same figure stops looking like a dose and starts looking like a supply calculation. That difference is the reason this site converts every figure rather than quoting it.
At the other end, a single 0.1% drop of 50 mcg is half a unit at the 1% strength, which is below what a U-100 syringe can read. The two ends of the same compound's figures therefore both fall outside what one syringe measures, at opposite edges.
What the figures omit
The trial figures carry a route — intranasal in every human study opened — a population that is usually neurological and often acutely ill, a course structure of 5 to 20 days rather than continuous use, and a manufactured solution of stated strength. They do not carry a subcutaneous dose, because no human study used that route; they do not carry pharmacokinetics in people; and outside Russia they do not carry a regulator. The mechanism and the evidence file are covered in Medibact's Semax guide, and the pairing most often discussed alongside it has its own page at Selank and Semax. Peptifact's Semax dosage page reads the same literature from the journalism side.
What this page does not do
It does not state an amount of Semax to use, a frequency, a course or a route. It converts percentages into mg/mL, mg/mL into micrograms per syringe unit, and reports which study or registration each figure came from, including one whose printed units cannot be right. Decisions about administering anything to a person belong with a clinician.
Sources and dates
Opened 2026-09-20: PubMed through the NCBI E-utilities for Semax in title or abstract (208 records), with the publication-type filters for clinical trial (4 records) and randomised controlled trial (1 record), and the abstracts of PMIDs 11517472, 9173745, 10741256, 18379501, 30225715, 29798983 and 27051926 read; the ClinicalTrials.gov v2 API for Semax as an intervention (no records); the DailyMed SPL search API for Semax (no records); PubChem compound record 9811102 for the sequence and molecular weight; MedlinePlus, "Giving an insulin injection" (reviewed 2024-07-21). The registered drop strengths and the 0.1% and 1% Russian registrations are as recorded in this site's dosage chart on 2026-09-18 from the Russian drug-reference entries. Corrections go to the contact page.
