What the Grey Market Costs

We track live inventory and pricing across roughly 31 grey-market vendors. Stripped to one number per molecule, here is what Chinese-origin peptides actually cost — not per vial, but per milligram of active, the only unit that lets you compare anything.

Figures are median prices across the 13 China-origin vendors that listed a standard 10-vial kit in the 30 days to June 2026 (9 to 12 vendors per compound). Methodology at the end. This documents an unregulated market; it is not an endorsement, and it is not dosing advice.

A vial price tells you almost nothing. A "$80 kit" of a 5mg peptide and a "$120 kit" of a 30mg peptide are not remotely comparable until you divide by the milligrams inside. So we did, across every recent China listing we hold, and reduced it to the going rate per milligram for the ten most widely stocked compounds.

The headline: this is now a commodity market, and it is priced like one. The blockbuster GLP-1s have collapsed toward cents per milligram, while the quiet research peptides — the ones that sound cheap by the vial — are what actually costs you. What the market actually charges:

The going rate, by molecule. Median price per milligram of active, computed across China vendors that listed standard 10-vial kits in the last 30 days. Bar is the median vendor; the thin line is the full vendor range. Per-mg flips the intuition: NAD+ and GHK-Cu look cheap because they are dosed in the hundreds of milligrams, while the peptides that sound cheap by the vial (tesamorelin, cagrilintide, TB-500) are the genuinely expensive ones to actually dose.

The GLP-1 price war

Tirzepatide is the cheapest serious drug on the board at a median of $0.40 per milligram. Retatrutide, the newer and more hyped GLP-1, still carries a premium at $0.77 per milligram — roughly 2 times tirzepatide for the same delivery mechanism. That gap is the price of novelty, the kind that erodes as more factories bring a compound online. Semaglutide, the oldest of the three, sits between them. None of this tracks dose accuracy: as the certificate data shows, the GLP-1s are middle-of-the-pack on whether the vial actually holds its label.

The spread is the tax

The single most useful number on this page is not the median — it is the range. The same molecule, from different China vendors, in the same month, varies by as much as 6x for Semaglutide, and a 3-to-5x range is normal across the board. There is no quality story underneath most of that spread; it is simply what a vendor thinks it can charge. Buying from the wrong storefront is a self-inflicted tax of several hundred percent, and it is invisible until you put the prices side by side.

Cheap does not buy you the dose

It is tempting to read price as a quality signal — you get what you pay for. The data says otherwise, flatly. The cheapest categories per milligram carry some of the worst underdose rates we measure, and the mid-priced GLP-1s are among the better-dosed. Price and dose-accuracy are independent axes. A low number per milligram buys you a low number per milligram; whether the labeled dose is actually in the vial is a separate question that only a third-party test answers. That is the whole case for testing.

Price is not a quality signal. If cheaper meant worse, the dots would climb left to right. They do not. The cheapest categories per milligram (GHK-Cu, the melanotans) carry some of the HIGHEST underdose rates, while the mid-priced GLP-1 blockbusters are among the better-dosed. Paying more does not buy you the labeled milligrams; only a third-party test does. (GLP-1 / metabolic in gold.)

What you actually get in the vial

Every price on this page assumes the label is honest — that a vial sold as 10 mg holds 10 mg. So we checked it against the tens of thousands of third-party certificates we hold, and the answer is more interesting than either the boosters or the doomsayers will tell you. The typical vial is not underdosed. If anything it is slightly overdosed: for every major compound the median tested result lands at or a little above the label, and the average runs a few percent higher still, because vendors pad their fills to stay ahead of complaints. On the median vial, per real milligram, you are not overpaying at all — you may even be getting a small bonus.

The cost is not in the average. It is in the variance.

The average is generous; the tail is the cost. Each row is the spread of third-party tested results for one compound, as a percentage of the labeled dose: the dark tick is the median, the box holds the middle half of vials, the thin line runs the 5th to 95th percentile. Two things stand out. The median sits at or above 100% — vendors overfill, so the typical buyer is not shortchanged — and yet every compound trails a left tail across the underdose line (shaded), from about one vial in eleven up to one in four. The number on the right is how often a compound lands more than 10% under label. Price does not move these distributions; only testing the vial in front of you does.

Underneath that generous median sits a real downside tail. Depending on the compound, between roughly one vial in eleven (9% for tirzepatide) and one in four (26% for ghk-cu) comes in more than ten percent under label — and when a vial is short, it is short badly, the underdosed ones averaging only about three-quarters of their stated dose. That is the real effective cost of a grey-market milligram: not a flat markup, but a lottery. Most draws pay out at or above face value; a meaningful minority short you by a quarter, and nothing about the price you paid tells you which vial you are holding. As the price-versus-quality scatter above shows, paying more does not buy a tighter distribution. The only thing that collapses the uncertainty to zero is testing the vial in your hand.

What the legal market charges

Three of these molecules are household names in a different bottle. Ozempic and Wegovy are semaglutide. Mounjaro and Zepbound are tirzepatide. These are not cousins of what the grey market sells — they are the same drug, and the distance between the two prices is the most revealing number on this page.

Start with semaglutide. Bought as Wegovy, self-pay direct from Novo Nordisk, it runs about $349 a month (cut from $499 in late 2025); as Ozempic the self-pay price is $349 to $499; with no program at all the retail cash price sits near $1,000. Compounded semaglutide from a telehealth pharmacy has settled around $99 to $149 a month. Now hold the dose at a standard 2.4 mg per week and price the same milligrams on the grey market: about $4.50 a month. That is roughly 78 times cheaper than the brand, for an identical molecule — and because semaglutide is dosed so low, its per-milligram gap is the widest on this page.

Tirzepatide tells the same story one notch less extreme, because it is dosed higher. Zepbound self-pay through Eli Lilly's LillyDirect is about $449 a month for maintenance vials; compounded runs $125 to $399; the grey market sells a month of the same 10 mg/week dose for roughly $16, about 28 times cheaper. Here is what a single month of the identical dose costs across three markets:

The same drug, three prices. Ozempic and Wegovy are semaglutide; Mounjaro and Zepbound are tirzepatide — the identical molecules priced on this page. Holding the dose fixed at a standard maintenance amount, a month of brand-name self-pay runs hundreds of dollars, compounded a fraction of that, and the grey-market raw material a few dollars. The gap is not subtle: on a log axis, where every gridline is ten times the last, the grey-market dot sits two to three full gridlines below the pharmacy shelf. Brand figures are manufacturer self-pay (NovoCare, LillyDirect) as of 2026; dose basis in Methods.

Insurance changes that number only if you can actually get it. When a plan covers these drugs the copay typically runs $25 to $150 a month, and a manufacturer savings card can push a commercially-insured patient's share toward zero. The catch is the coverage itself. Ozempic is usually covered for type-2 diabetes, but for weight loss only about 45% of large employers cover any anti-obesity drug (up from a quarter in 2023); Medicare excluded obesity drugs outright until a temporary 2026 bridge program at $50 a month; and prior-authorization denials are routine. That gap — a drug that works, priced for the few with coverage and list-priced for everyone else — is the vacuum the cash, compounded, and grey markets all rushed to fill.

Retatrutide is the exception that proves the rule. It has no legal price at all, because it is not approved — still in late-stage trials, available nowhere with a prescription. For reta the grey market is not the cheap option, it is the only option, which is much of why the hype, and the markup over tirzepatide, runs as hot as it does.

None of this discount is free. The legal price buys a finished, sterile, identity-verified injectable made under inspection, a labeled strength you can trust, and a prescriber who is accountable for it. The grey-market price buys raw lyophilized powder in an unlabeled-strength vial, no oversight, and — as our certificate data shows — a coin flip on whether the milligrams on the label are actually inside. The discount is real, but it is a risk premium you are choosing to carry yourself, and the only way to convert any of it back into safety is to test what actually arrived.

$0.40

per mg, tirzepatide

the cheapest GLP-1 to dose

$0.77

per mg, retatrutide

~2x tirzepatide

78x

cheaper, same semaglutide

grey market vs Wegovy self-pay

28x

cheaper, same tirzepatide

grey market vs Zepbound self-pay

price spread, semaglutide

same molecule, vendor to vendor

China vendors priced

active in the last 30 days

Methods and limitations

Source. Every grey-market figure is computed live from our reconciled vendor pricelist table (deduped, kit-corrected, freshest-live-stock) filtered to origin_country = CHINA, standard 10-vial kits, listed in the last 30 days, across the 13 China vendors active in that window (of roughly 31 we track in total). Price per milligram = kit price / (10 vials x milligrams per vial). Listings dosed in IU or by volume, and multi-peptide blends, are excluded because they have no clean single-molecule milligram figure.

Why median, not average. We report the median, and it matters. Grey-market price data is right-skewed and pocked with extraction errors — a single mistyped or mis-parsed listing (a kit price recorded as a single vial) can read as ten times the real rate, and an average swallows that whole. The median does not move. Concretely, before cleaning, one tesamorelin row dragged the mean per-mg above $2.80 while the median sat near $1.80; the median was right. So we (1) drop any offering more than 3x off the compound's raw median as a parse error, (2) take each vendor's own median first, so a vendor listing many doses cannot dominate, then (3) report the median across vendors. After that cleaning the mean and median nearly converge, which is the signal that the noise is gone — but we keep the median as the headline because it answers the only question a buyer has: what does this typically go for. The mean answers a question nobody asked.

Legal-market reference prices are manufacturer self-pay and telehealth-compounded figures verified for 2026: Wegovy and Ozempic self-pay via Novo Nordisk's NovoCare (Novo cut the standard Wegovy self-pay price from $499 to $349 a month in November 2025); Zepbound self-pay single-dose vials via Eli Lilly's LillyDirect "Self Pay Journey" ($299 / $399 / $449 by dose, effective February 2026); and 503A compounded ranges reported across telehealth pharmacies. They are converted to a per-month and per-milligram basis at a fixed maintenance dose — semaglutide 2.4 mg/week, tirzepatide 10 mg/week, four weekly doses to the month (the unit the self-pay price is actually sold in) — so the comparison is the same molecule and the same milligrams across two markets. Insurance copay and coverage figures (a typical covered copay of $25 to $150 a month; roughly 45% of large employers covering an anti-obesity drug in 2026; Medicare's temporary $50-a-month obesity bridge) are reported 2026 ranges and vary widely by plan and indication. They are a different product class (finished, sterile, prescriber-included injectables) and are shown for scale, not as a like-for-like swap.

What this is not. A median is a snapshot of asking prices, not a recommendation, and a low price is not a safety claim. Cross-reference every number here against the dose-accuracy and purity data before it means anything. Every grey-market figure on this page regenerates directly from live vendor inventory; the queries are reproducible.