In mice, a tiny dose of semax (about 0.05 mg per kg body weight) helps them cope with low‑oxygen conditions, and a short 6‑day course at 0.1 mg/kg does the same. It also improves memory in several tests, working about as well as common nootropics like piracetam, but the benefit follows a bell‑shaped curve—too low or too high a dose loses the effect. Semax doesn’t boost brain oxygen use like mexidol, so dosing must be precise.

Upon single administration, mexidol and semax only in doses of 100 and 0.05 mg/kg, respectively, produced an antihypoxic effect on mice in the altitude chamber and hermetic chamber tests. Preventive course administration of mexidol and semax for 6 days gave significant antihypoxic effect on the model of acute hypobaric hypoxia in mice in doses of 75 and 0.1 mg/kg per day, respectively, in which the same preparations upon single administration were ineffective. Neither mexidol nor semax upon single administration were effective on the models of acute hemic and histotoxic hypoxia. On various models of amnesia (except that induced by the maximal electroshock) in mice, both mexidol and semax exhibited marked antiamnesic effects comparable with that of the reference nootrope drugs piracetam and oxyracetam. Mexidol showed a linear, while semax exhibited a bell-shaped reversible dose-effect relationships. Mexidol and semax inhibited the ortho- and antidromic population response spikes of CA1 pyramidal neurons of survival hippocampal slices in rats. It was estimated that mexidol (in contrast to semax) increased oxygen consumption in rat brain mitochondria and had a linear dose-effect relationship in a concentration range of 1-5 mM. It is concluded that mexidol should be used in high doses (for both single and course administration) for obtaining antihypoxic and antiamnesic effects, while semax requires a thoroughly controlled choice of dosage.