GLOW Dosage Guide 2026: What Is Actually In the Vial

What is GLOW?

A vendor name over a mixture: GLOW is a label for a pre-mixed vial of three separate peptides, sold for research use. It is a mixture with a name, not a molecule.

What the 70 means: the vial is written as GLOW-70, GLOW 70, GLOW 70mg or glow70, and the number is the total dry milligrams of all three peptides in the vial before any water goes near it. It is not an amount per injection, not a strength per mL, and not a per-component figure. On the label the vial carries, those 70mg break down as GHK-Cu 50mg, BPC-157 10mg and TB-500 10mg.

The spellings, all of them one vial: glow, glow 70, glow-70, glow 70mg, glow blend, glow protocol, glow stack and glow peptide name the same product. The punctuation is noise, the word peptide is noise, and the 70 is the label total every variant shares.

Why "compound" is the wrong word: a compound carries a single sequence, a single formula, a single mass and a single registry entry. A three-peptide blend carries three of each. GLOW has no CAS number of its own, no molecular weight of its own and no database entry of its own, because there is nothing single there to register.

GLOW-70 is not KLOW-80: the two names are one letter apart and the KLOW vial carries all three of GLOW's components, which makes them easy to confuse. KLOW is labeled 80mg, the same three peptides plus KPV 10mg, so the extra letter is an extra peptide rather than a different strength of the same thing, and our KLOW guide breaks that vial down component by component.

What GLOW is sold to do: the claims made for the blend, and what has been measured against each of them in a human being, are set out in claimed benefits against measured evidence further down.

Why there is no GLOW dosing chart on this page

The chart exists, and we cannot show you whose it is: tables putting an amount against the 70mg vial are not hard to find, and each one we located had been put up by somebody in the business of shipping that vial. Two rules meet here: a source gets named or the claim goes, and a seller never gets named. Between them there is nothing left of those tables to reprint, and nothing left to launder into a table of our own.

The version of this page we refused to write: keeping the volumes and the amounts, hanging them on the phrase "a published chart", and letting the grammar of attribution do the work a link is supposed to do. Nothing in that sentence resolves to a document. Attribution with no name behind it is not a thinner citation, it is the absence of one dressed up as the presence of one, and this library was built to make that impossible rather than merely rare.

The cost, itemized: four questions go unanswered here. How much water joins the 70mg of powder. What quantity of the solution that produces counts as one injection. How much time sits between injections. How many weeks a run lasts. Answers to all four are in circulation, none of the four has an origin we can hand to a reader, and so all four are left standing open rather than borrowed.

What needs nobody's permission: the arithmetic. Concentration is 70 divided by however much water goes in, and what a single unit carries falls out of that concentration together with whichever of the two syringe scales is being read. Both steps run below on five volumes we chose ourselves, to show the shape of the curve rather than to point at a number.

Behind the frequency on any of those tables: nothing measured. A 2026 narrative review of injectable peptide therapy in the American Journal of Sports Medicine, covering BPC-157, thymosin beta-4, TB-500 and GHK-Cu among others, concludes that the indications, dosing, frequency and duration of treatment for these peptides all remain unknown (Mayfield et al., 2026). That verdict is about the components on their own. For the three-peptide mixture there is not even that.

We will not pick a number for you, and no page that has never been asked to show its work is in a position to either.

The question a chart could not settle even if we had one: whether three peptides sharing a vial, in a proportion no standard enforces, do anything the three would not have done out of three separate vials. That is a question for an experiment, no experiment on the mixture has been run, and stating a schedule more confidently answers none of it. The general case, one vial producing two honest and incompatible charts, is worked through in why two dose charts disagree.

The GLOW dosage calculator

Whole-blend arithmetic, not per component: it takes a vial size, a water amount, a syringe type and an amount in milligrams, and returns the concentration, the volume and the unit reading for all three peptides together.

GLOW reconstitution: the 70mg vial at each water volume

The powder half is already settled: 70mg is what the label commits to, which leaves the water as the only variable in the room. The five rows below are volumes we picked ourselves. Not one of them is a recommendation, not one was lifted out of anybody's schedule, and the table nominates no row as the correct one.

The line worth taking from that table: the vial has no concentration at all until someone decides the water. Every unit count downstream of that decision, on every chart and in every calculator, is a consequence of a choice that never gets written on the glass.

Once the water is in, how long does it keep: we have no citable answer. Shelf lives for the 70mg vial are quoted freely, every quotation we traced ended at a seller, and that puts all of them outside what this page will repeat. The one detail here that is chemistry rather than a stability claim is the color: GHK-Cu carries a copper ion, the dry powder is blue, and the solution takes on a blue-green tint. That holds before and after the water goes in, and it settles nothing about duration.

How does the units math work?

What 70mg on a label buys you: nothing, until someone decides the water. Every volume named from here on is one we chose to make the arithmetic visible, and no figure below is attached to a person, a body or a schedule.

A 70mg vial at two milliliters, worked out: a 70mg GLOW vial holding 2mL of bacteriostatic water sits at 35mg per mL, or 35,000mcg per mL.

Thirty-five hundredths of a milligram to a unit: the U-100 scale puts 100 units in a milliliter, so at that concentration one unit carries 0.35mg of blend.

The same vial at three milliliters: 23.3mg per mL, with a unit carrying 0.233mg rather than 0.35mg.

And at the outer edges: 1mL leaves the vial at 70mg per mL and a unit at 0.7mg. 5mL leaves it at 14mg per mL and a unit at 0.14mg, a fifth of what the same unit carried at 1mL.

Which is the whole point: the unit reading did not change, and what it carries moved by a factor of five. Two people converting the same unit count off the same 70mg vial can be handling masses five times apart, and nothing about the syringe records which of the two they are holding.

And a third input the water argument never reached: a unit is 0.01mL on one of the two printed scales and 0.025mL on the other, sitting behind the label total and the water volume this section has already named.

The blend wrinkle: at the 50/10/10 label split, 71.4% of whatever mass a draw carries is GHK-Cu and 14.3% is each of BPC-157 and TB-500, whatever the size of the draw. Converting units moves the 70mg total around. It cannot move those three percentages, which were settled when the vial was filled.

So a blend leaves four things off the syringe, not two: the water that set the concentration, the scale printed on the barrel, the total claimed on the label, and the split hiding inside that total. The syringe never lies about volume. It simply has no access to any of the four.

What is a GLOW protocol, and how long is a cycle?

Two meanings of one word: in a trial, a protocol is a document filed before the first subject is treated, and the study is then run to it and reported against it. Around a 70mg vial the same word covers something with none of that behind it: an amount, a spacing and a count of weeks, typed by somebody, answerable to nothing.

So no schedule and no run length appear here: every week count we found attached to this vial had a seller behind it, and a seller is precisely what this site will not put in front of a reader as a source. A run length nobody can trace borrows an authority it was never granted, and passing one along would do more damage than leaving the question standing open.

What the literature has on duration: nothing measured. The same 2026 narrative review quoted above puts duration of treatment on its list of unknowns for these peptides (Mayfield et al., 2026), and that verdict is about the components. The mixture has no trial, so it has no duration to be unknown about.

The one thing left, and it is only a shape: a weekly total is a per-injection amount multiplied by a frequency, and the number of weeks one 70mg vial covers is 70 divided by that weekly total. Two inputs, neither of them with a source we could hand over, so the expression stays an expression and never becomes a number.

The honest answer to "how long is a GLOW cycle": the study that would answer it has not been run. A count of weeks only becomes a finding once something was measured at the end of it, and for this three-peptide mixture nothing has been measured at any point.

Do the GLOW charts change for women, for beginners, or for bodyweight?

The direct answer: no, and nothing on this page is built to. Sex, bodyweight, age and training history appear in no column of any table here, for the plain reason that there is no measured population to split on and no document we could cite if we split anyway.

Why there is nothing to condition on: a figure that changes by population has to come from a population somebody measured. FDA reports no clinical studies or human exposure data for TB-500 by any route, limited data in humans for injectable GHK-Cu, and for BPC-157 a short list of small studies by routes other than the one this blend is sold for (FDA, PCAC July 2026).

What a beginner chart would be: the same numbers wearing a softer label. A smaller number is still a number somebody chose, and labeling it a starting point attaches no evidence the larger one lacks.

So this page prints no figure of its own, for anyone: not by sex, not by first vial or tenth, not by how much a body weighs. The converter above changes the units on a figure you already have. Where that figure came from is the question this whole page exists to ask.

Why GLOW is a blend and not a stack

The distinction people are searching for: "glow stack" and "glow blend" return the same product, but the two words describe opposite things. A stack means separate vials, each drawn on its own, with the ratio between them settled at the syringe every time. A blend means one vial in which that ratio was fixed before it shipped and cannot be moved after. GLOW is the second kind, whatever the search box calls it.

What the ratio is: on the 50/10/10 label, the 70mg vial is 71.4% GHK-Cu and 14.3% each of BPC-157 and TB-500. Those three percentages ride along on every draw, large or small, so half a draw is half of each of the three and twice a draw is twice each of them.

That fixed ratio has an arithmetic consequence, not a debatable one: there is no draw from a GLOW vial that moves one component without moving all three in lockstep. Halving the copper peptide halves the BPC-157 and the TB-500 at the same moment. Three separate vials can be adjusted one at a time; a blend cannot, by construction. The general form of the problem is set out in our note on doing the math on blend vials.

Where the mass actually sits: GHK-Cu is 50 of the 70 milligrams, so most of what a GLOW syringe carries is the copper peptide, and most of whatever any effect or any side effect would have to be attributed to is the copper peptide too. The two 10mg components are a seventh of the vial each.

A label claim, and the certificate we are not able to show you

The 50/10/10 split is a label claim: it is what the vial is sold as. It is not a standard, nothing enforces it, and there is no monograph, no compendial specification and no regulator that has ever defined what a GLOW vial must contain. The 71.4% and the 14.3% above are therefore arithmetic performed on a claim, not on anything anyone weighed.

There are certificates of analysis for the 70mg vial, and none of them is cited here: each one we located sits on the site of somebody selling the vial, so linking it means naming that seller, which is the one move this site does not make. The certificate stays unshown, and an unshowable document is not one we will lift assay figures out of. The shape of the problem survives; the numbers inside it do not. Our write-up of why a purity figure and a content figure answer different questions is in what an HPLC purity figure does not tell you.

A single-peptide vial asks for four acts of trust, and this one asks for five: that the contents are what the label names, that they sit in the stated proportion, that they are pure, that they are sterile, and that 70 is the real dry mass. The milligrams can be wrong while the ratio is right, and the ratio can be wrong while the milligrams are right, which is two ways to fail rather than one, and the unit counts above need both of them to have gone the other way. We can point at nothing that tests either. The closest thing to a measurement anywhere in our sources sits in what is not known yet below, and it describes the supply as a whole rather than the vial in anyone's hand.

GLOW peptide benefits: the claim, and what was measured

Why this arrives as a table rather than a list of benefits: a benefits list would only tell you what the sellers say. The interesting column is the second one, which asks which component a claim is resting on, and the third, which asks what anyone has measured in a human being.

The regulator's version of the same question: when FDA reviewed two of these three for the 503A compounding list in July 2026, it recorded the use each had been nominated for. BPC-157 was evaluated for ulcerative colitis, and the agency concluded there was a lack of evidence to support its effectiveness there. TB-500 was evaluated for wound healing, with the nominator not specifying what kind of wound, and the agency concluded there was a lack of evidence to evaluate effectiveness at all. On both it proposed that neither the free base nor the acetate form be added to the list.

What that does and does not mean: "not enough evidence" is not the same sentence as "does not work". It means the study has not been run. Only one of those two sentences is being sold to you, and it is not the one the agency wrote.

GLOW before and after, and how long it takes to work

The direct answer: there is no onset figure for GLOW, because no trial of the blend has measured a time course. Week-by-week timelines for this mixture do circulate, skin texture at one range of weeks and collagen at another, and none of them is a readout from anything that was measured.

Why a before-and-after is weak evidence here specifically: a before-and-after needs a comparator, and photographs have no control group, no blinding and no record of what was in the vial. That last one is not a technicality on a blend: nothing on the glass, and nothing this page can show you, establishes how many milligrams of which peptide were in the solution being photographed.

And the effect being photographed has a named competitor: a 2026 review of peptide use in sports medicine covering BPC-157, GHK-Cu, thymosin beta-4 and TB-500 among others notes that many of these compounds show favorable repair outcomes in animal models while rigorous human safety data stay scarce, and it discusses the placebo effect as a mediator of peptide efficacy, amplified by social media (Mendias and Awan, 2026). An uncontrolled recovery story is the exact shape of evidence that discussion predicts.

Where the side-effect question is answered: there is no trial safety profile for this mixture, so there is no side effect table on this page. What the regulator's own adverse event file records is in the safety picture below. Three voluntary reports, every one of them filed against BPC-157, is the whole of it, and three reports touching one of the three peptides is not a profile of the vial.

What's not known yet?

Why an approval is not even possible here: GLOW is not FDA approved, none of its three components is approved for general use, and the blend could not be. There is no single manufacturer, no enforced ratio and no trial to file.

The regulator's file: all three components sit in the "nominated but withdrawn" table on FDA's compounding safety-risk page, which lists substances previously in category 2 of the interim policies whose nominations the nominators withdrew. We read that page on 2 August 2026 and it dates its own content 22 April 2026. For TB-500 it states the agency has not identified any human exposure data. For injectable GHK-Cu it cites limited data in humans. For BPC-157 it states the agency has identified no, or only limited, safety-related information for the proposed routes.

What happened on 23 July 2026: BPC-157 and TB-500 went before FDA's Pharmacy Compounding Advisory Committee, considered for the 503A Bulks List alongside KPV and MOTS-c. The nominations had been withdrawn by their nominators and the agency elected to present them anyway. In its briefing document FDA proposed that BPC-157 free base, BPC-157 acetate, TB-500 free base and TB-500 acetate all be kept off the list.

The committee voted against the agency's own scientists: the Associated Press, reporting from the meeting on 23 July 2026, records the panel voting 8 in favor, 6 against and 1 abstention across a series of votes, advising the agency to drop the restriction that bars pharmacies from producing injectable peptides including BPC-157 and TB-500. NBC News, reporting the same day, records that eight members of the panel had been recently appointed under the health secretary, six of them running clinics that offer peptides. The recommendation is advice only and the agency is not bound by it, since the decision comes through rulemaking. As of 2 August 2026 the FDA meeting page publishes an agenda, briefing documents and slides, with no minutes, no vote record and no decision.

The blend-shaped holes:

• No study of the mixture at any amount. The PubMed query recorded in the key figures above returns four records on 2 August 2026, and all four are narrative reviews of the individual peptides.
• No registered trial of the mixture either. ClinicalTrials.gov returns nothing for GHK-Cu and BPC-157 together, searched 2 August 2026.
• No data on how the three components behave once combined in one solution.
• No enforced ratio: the 50/10/10 split is a label claim rather than a standard, and nothing enforces it.

Evidence for a component is not evidence for a mixture, and the mixture has none of its own.

What would change our read: a controlled human trial of the mixture, which does not exist and is not registered. Two of the three components do carry small controlled human studies, and the pair of them share a limitation: both were applied to skin. In the GHK-Cu one, a skin care regimen used after carbon dioxide laser resurfacing, the blinded objective measures separated the groups by nothing across 13 completers, and the only endpoint reaching significance was how the patients themselves rated their overall skin quality, at P equal to .04 (Miller, 2006). In the thymosin beta-4 one, a double-blind, placebo-controlled, dose-escalation phase 2 in 73 randomized patients with venous ulcers, the safety profile came out comparable to placebo (Guarnera, 2010). No injection in either, and no mixture in either. Once both are set aside, what is left standing is a narrow claim about three peptides sharing one vial and nothing beyond it.

The market the vial comes from: a 2026 preprint, not peer reviewed, analyzed 6,441 peptide samples across fourteen compounds including BPC-157, GHK-Cu, TB-500 and thymosin beta-4, and reports that between 41.6% and 71.1% failed basic quality criteria depending on which of two acceptance models was applied, with measurable endotoxin in 15% of samples. That is a statement about the supply these vials come out of, not a measurement of any particular vial, and it is as close to the question of what is in the glass as this page can honestly get.

The blend-specific version of the problem: a vial holding one peptide can be wrong about its milligrams. A blend can be wrong about its milligrams and wrong about its ratio at the same time, and every unit count on this page needs both of those to be right. There is no independent check on either that we can put in front of a reader, which is the honest end of that sentence rather than a reassurance.

How this page is sourced

None of this passed through a summary: the research figures came off the cited abstracts and the regulatory claims off FDA's own pages and meeting documents, never off anyone's summary of either.

The eleven sources: two randomized human trials, three reviews of a single component, two 2026 reviews of the class, one 2026 preprint, two regulator records, and one entry for the news reporting of the committee vote, with DOIs, PubMed IDs and links. Each is listed below with its identifier, and every figure on this page traces back to one of them.

What changed on 2 August 2026, and what it cost: until today this page compared what four published dosing charts state for the same vial, and quoted three certificates of analysis for a 70mg vial. All seven of those documents are published by sellers of the blend. Under the rule this site now works to, a source we cannot name is a source we cannot cite, and a claim we cannot cite is cut rather than softened. So the chart comparison went, and with it every water volume attributed to anyone, every per-injection amount, every weekly total, every cycle length, the storage window, and every assay figure, together with the seven reference entries that carried them.

Four of those entries were added earlier the same day, to fix the opposite problem: the chart pages had been named throughout the page above a reference list that held none of them. Adding the entries was right under the rule as it then stood and is wrong under this one. Citing a source properly and not naming it at all cannot both be satisfied, and when they collide the claim goes rather than the rule.

One registry record went the same day, for a related reason: this page carried NCT07437547 as a live randomized phase 2 of BPC-157, 120 participants planned, and called it the one study on the register worth following. The sponsor named on that record has eight studies on ClinicalTrials.gov, two of which describe themselves as specimens in their own summaries, one opening "This fictional study is an example of a ClinicalTrials.gov-style record". All eight share one facility and one pair of contact addresses, on an email domain that does not match the sponsor's name. We cannot show that NCT07437547 is a fabrication and we are not claiming it. We are saying it cannot carry a sentence about what is coming, so the sentence is cut and the paragraph now says what is true: no trial of the mixture exists and none is registered.

What we did not do: keep the numbers and attribute them to "a published chart", to "community charts" or to "a chart we read". That reads as sourced while pointing at nothing a reader can open, which is worse than saying less, because it looks like evidence.

What is left is what we can show you: the label the vial carries, the arithmetic that follows from it, the published evidence on each of the three peptides one at a time, and the regulator's own record. Where a question has no answer we can source, this page now says so instead of answering it.

The bar a citation has to clear before it goes in: that bar is written out on the methodology page, together with how a correction to any figure here is logged.

Last reviewed: 2 August 2026.

1. 1

   Miller TR, Wagner JD, Baack BR, Eisbach KJ. Effects of topical copper tripeptide complex on CO2 laser-resurfaced skin. Arch Facial Plast Surg. 2006;8(4):252-259. doi:10.1001/archfaci.8.4.252. PMID 16847171.

2. 2

   Guarnera G, DeRosa A, Camerini R. The effect of thymosin treatment of venous ulcers. Ann N Y Acad Sci. 2010;1194:207-212. doi:10.1111/j.1749-6632.2010.05490.x. PMID 20536470.

3. 3

   Józwiak M, Bauer M, Kamysz W, Kleczkowska P. Multifunctionality and Possible Medical Application of the BPC 157 Peptide: Literature and Patent Review. Pharmaceuticals (Basel). 2025;18(2):185. doi:10.3390/ph18020185. PMID 40005999.

4. 4

   Gwyer D, Wragg NM, Wilson SL. Gastric pentadecapeptide body protection compound BPC 157 and its role in accelerating musculoskeletal soft tissue healing. Cell Tissue Res. 2019;377(2):153-159. doi:10.1007/s00441-019-03016-8. PMID 30915550.

5. 5

   Goldstein AL, Hannappel E, Sosne G, Kleinman HK. Thymosin beta-4: a multi-functional regenerative peptide. Basic properties and clinical applications. Expert Opin Biol Ther. 2012;12(1):37-51. doi:10.1517/14712598.2012.634793. PMID 22074294.

6. 6

   Mayfield CK, Bolia IK, Feingold CL, Lin EH, Liu JN, Rick Hatch GF, Gamradt SC, Weber AE. Injectable Peptide Therapy: A Primer for Orthopaedic and Sports Medicine Physicians. Am J Sports Med. 2026;54(1):223-229. doi:10.1177/03635465251357593. PMID 41476424.

7. 7

   Mendias CL, Awan TM. Safety and Efficacy of Approved and Unapproved Peptide Therapies for Musculoskeletal Injuries and Athletic Performance. Sports Med. 2026 Apr 12. Online ahead of print. doi:10.1007/s40279-026-02437-0. PMID 41966639.

8. 8

   U.S. Food and Drug Administration. Certain Bulk Drug Substances for Use in Compounding that May Present Significant Safety Risks (Category 2), including the "nominated but withdrawn" table that carries BPC-157, GHK-Cu for injectable routes of administration, and Thymosin beta-4, fragment (LKKTETQ), also known as TB-500. Content current as of 22 April 2026. Retrieved 2 August 2026, HTTP 200. fda.gov/drugs/human-drug-compounding.

9. 9

   U.S. Food and Drug Administration. July 23-24, 2026 Meeting of the Pharmacy Compounding Advisory Committee: meeting notice and agenda; FDA Briefing Document, whose Points to Consider propose that BPC-157 free base, BPC-157 acetate, TB-500 free base and TB-500 acetate not be included on the 503A Bulks List; and the FDA presentations delivered on 23 July 2026, which carry the BPC-157 human exposure list, the three adverse event reports, and the finding that no information was identified in which TB-500 had been given to patients. Retrieved 2 August 2026, HTTP 200. meeting page, briefing document, 23 July presentations.

10. 10

    News reporting of the vote. This is not an FDA record. Perrone M. Associated Press, 23 July 2026, reporting the panel voting 8 in favor, 6 against and 1 abstention across a series of votes on BPC-157, TB-500 and KPV. apnews.com. Lovelace B Jr, Miller SG. NBC News, 23 July 2026, recording that eight panel members had been recently appointed under the health secretary and that six of them run clinics offering peptides. nbcnews.com. Both retrieved 2 August 2026, HTTP 200.

11. 11

    Preprint, not peer reviewed. Mendias CL, Awan TM. Evaluation of Research Grade Peptides Marketed Directly to Consumers Reveals Extensive Variability in Purity and Measured Abundance. Preprints.org, posted 24 April 2026: 6,441 samples across fourteen compounds including BPC-157, GHK-Cu, TB-500 and thymosin beta-4, 41.6% to 71.1% failing quality criteria depending on the model applied, measurable endotoxin in 15%. doi:10.20944/preprints202604.1748.v1. Record and abstract confirmed through CrossRef and Europe PMC on 2 August 2026; the publisher page returned HTTP 403 to an automated request.