NMNH is NMN carrying two extra hydrogens — the reduced form of the same molecule. In cells and in mice it raises NAD⁺ higher and faster than NMN does, so the “more potent” claim on the label is, for once, describing something real. It has never been given to a person in a published trial.
That combination is the whole story: a genuine laboratory result being sold as a finished product, five years early.
What the reduced form actually does differently
The molecule was first reported as an NAD⁺ precursor in 2021, when a group synthesized it and found it raised NAD⁺ to a much higher extent, and faster, than either NMN or nicotinamide riboside. The interesting part was the route: NMNH is metabolized through a different pathway, independent of NRK and NAMPT — the enzymes the conventional precursors depend on. In mice, a single administration produced a rapid and sustained NAD⁺ surge in whole blood, with raised NAD⁺ in liver, kidney, muscle, brain, brown adipose tissue and heart, though not in white adipose tissue [1].
A second 2021 group, working independently with its own synthesis route, agreed on the headline: NMNH was a better NAD⁺ enhancer than NMN both in cells and in mice, mediated by the enzyme NMNAT [2].
Bypassing the enzymes that normally rate-limit the conversion is genuinely novel chemistry, and it is the reason NMNH exists as a product at all. It is also the reason to be careful with it — a reduced molecule is not simply a stronger version of an oxidized one, a distinction this category also blurs between NAD⁺ and NADH.
The sentence that is not on any label
That second paper measured more than NAD⁺. NMNH also raised NADH — the reduced partner — in cells and in mouse livers, and metabolomic analysis found it inhibited glycolysis and the TCA cycle, the two central pathways that turn food into usable energy. In cell experiments it induced cell cycle arrest and suppressed cell growth [2]. The authors noted no observable difference in the mice’s weight.
Read that against the marketing and the problem is plain. NMNH is sold on the promise of more cellular energy. The published characterization of it reports suppression of the pathways that produce cellular energy, in the same experiments that establish its potency. Those are not two separate findings to be weighed against each other — they are the same property, seen from two sides. Shoving a large amount of reduced cofactor into a cell changes its redox balance, and a cell that has had its redox balance changed does not necessarily thank you.
None of this makes NMNH harmful in a person. It makes it uncharacterized in a person, which is a different and more honest thing to say.
The census, and the acronym problem underneath it
We searched PubMed in September 2026. A search for “reduced nicotinamide mononucleotide” in the title or abstract alongside trial, participants or volunteers returns zero records. Adding the acronym and filtering for the randomized controlled trial or clinical trial publication types, or for randomized or placebo in the text, also returns zero.
The control search run the same day is the one that makes those zeroes mean something: plain nicotinamide mononucleotide, filtered to the randomized controlled trial publication type, returns 20 records. The filter works. The reduced form simply is not in there.
One warning for anyone repeating this search. The bare acronym is close to useless, because NMNH is also the Smithsonian’s National Museum of Natural History: a search for NMNH alongside human returns fourteen records, and the recent ones are about paleopathology archives and primate specimens. Anybody citing an “NMNH literature” count without spelling the molecule out has probably counted museums.
What NMN has that NMNH does not
A bar to clear, and it is not a high one. In a randomized, double-blind, placebo-controlled, dose-ranging trial of 80 healthy middle-aged adults (n = 80), blood NAD rose significantly in every NMN group — 300, 600 and 900 mg daily — at day 30 and day 60 against both placebo and baseline (all p ≤ 0.001), and six-minute walking distance was significantly higher than placebo in all three dose groups (all p < 0.01) [3].
That is what a human dose-response looks like. For NMNH there is no equivalent — no published pharmacokinetic curve in people, no established dose, and therefore no relationship between the milligram figure on a bottle and anything measurable in your blood. A dose you cannot connect to an effect is a number, not a dose, which is also why the NMN dosing question has an answer and the NMNH one does not.
So should you buy it?
Not yet, and the reason is narrow enough to state without hedging: the only published characterization of what NMNH does to a cell includes suppressed growth and inhibited central metabolism, and nobody has followed that up in a human being. When a molecule’s selling point and its most concerning finding are the same mechanism, the missing trial is not a formality.
If what you want is a precursor with human data behind it, that is NMN or nicotinamide riboside, and which of those two has the better evidence is a real question with a real answer. If you want to know what any of them have actually been shown to do for you rather than for a blood marker, that is the evidence page, and the answer is less than the category implies. Prices for the precursors we have verified are on the supplement board. For plain NMN, which is what almost everyone is actually choosing between, the value ranking is here.