Educational reference only. Not medical advice. Research compounds are not for human or veterinary use.

Grey-market vs. compounded peptides: what's really different

You will often hear that research peptides and compounding-pharmacy peptides are basically the same thing: the raw material all comes from China anyway, and "research only" just means less paperwork. There is some truth in that, and some important gaps. Here is what the evidence actually shows about where peptides come from and what separates the two channels.

Where the raw ingredients come from

Every peptide product starts with an active pharmaceutical ingredient, or API: the bulk powder of the peptide itself. Most of it is made outside the United States.

  • Not all of it comes from China. In 2019 the FDA reported that, of the facilities making APIs for the U.S. market, 28% were in the United States, 26% in the European Union, 18% in India and 13% in China. The FDA also said it could not determine how much API China actually produces or how much of it reaches U.S. patients.
  • India and China dominate new filings. According to the U.S. Pharmacopeia, India accounted for 50% of new API drug master files (a type of manufacturing dossier filed with the FDA) in 2023 and China for 32%.
  • For peptides, China looms larger. Reuters reported that at least eight Chinese companies known to supply compounding pharmacies shipped enough semaglutide and tirzepatide API in 2024 for more than a billion starter doses.
  • The weakest links serve compounders. In its 2024 report on pharmaceutical quality, the FDA said API sites that supply only compounding pharmacies make up just 18% of API manufacturers in its catalog but 72% of the API sites hit with FDA regulatory action over the past five years. Those problem sites were mostly in China (51%) and India (30%).

So the accurate version of "it all comes from China" is this: China and India make most of the world's API, Chinese companies are major suppliers of peptide API, and compounding pharmacies and research-chemical sellers can end up buying from the same region and sometimes the same factories.

A real example of the overlap

In May 2026 the FDA sent a warning letter to Harbin Jixianglong Biotech, a Chinese maker of semaglutide and tirzepatide API. Among the findings:

  • some API was labeled "Only for R&D use purpose," but the FDA said the quantities shipped to the U.S. were inconsistent with research use;
  • the company bought semaglutide from other suppliers, relabeled it under its own name and shipped it, including material from an establishment not registered with the FDA;
  • it had shipped API without process validation, and without verifying its endotoxin test.

That one letter shows both sides of the argument. The same factory can sit behind both channels, and a "research" label upstream does not tell you much about quality.

What separates the two channels

The difference is mostly not the molecule. It is everything that happens to it, and everyone who has to answer for it.

Compounding pharmacies

  • 503A pharmacies make medicines for an individual patient with a valid prescription, under a licensed pharmacist and the oversight of state pharmacy boards.
  • By law, the bulk ingredient must come from a manufacturer registered with the FDA and arrive with a valid certificate of analysis. It must also be an allowed substance: one with a USP monograph, one that is part of an approved drug or one on the FDA's approved bulks list.
  • 503B outsourcing facilities register with the FDA, must follow current good manufacturing practice and are inspected by the FDA on a risk-based schedule.
  • State boards have said plainly that "research use only" material may not be used for compounding.
  • In late October 2026, HHS announced it would not take action against 503A pharmacies compounding six peptides, including BPC-157, TB-500 and MOTS-c, while rulemaking continues. Its conditions include API from an FDA-registered manufacturer that follows good manufacturing practice, lot testing documented on a certificate of analysis and a patient-specific prescription. Products sold as research chemicals are excluded.

None of this makes a compounded drug FDA approved, and the FDA does not check compounded drugs for safety or effectiveness before they are dispensed. But there is a named pharmacist, a prescriber, a paper trail for the ingredient and a regulator who can inspect.

Research-use-only sellers

A research peptide is sold for laboratory work, not for people. No law requires its seller to use a registered API source, to test each lot, to fill vials under sterile conditions, to check for endotoxin or to answer to a pharmacy board. Some sellers do test, and publish certificates of analysis. Many do not, and the buyer has no independent way to know which is which without testing the vial.

The FDA has also made clear that a "research use only" label is not a legal shield. In August 2026 it sent warning letters to five online peptide sellers, saying their websites, including dosing guidance, reconstitution calculators and bacteriostatic water sold alongside peptides, showed the products were intended for human use.

What independent testing has found

There is still very little published testing of grey-market peptides, and most of it concerns GLP-1 weight-loss compounds. What exists is not reassuring:

  • Online semaglutide (2024): Researchers bought three research-use-only semaglutide vials. Each contained 29% to 39% more semaglutide than labeled, all three contained endotoxin, and measured purity was far below the 99% claimed, which the authors attributed to unidentified impurities. The vials were sterile. Three attempts to buy Ozempic pens without a prescription were scams that never delivered.
  • Retatrutide in Australia (2026): Of 18 grey-market retatrutide samples tested by one Australian lab in 2026, only one matched its label. About 30% were stronger than labeled, about 30% weaker, and one contained no retatrutide at all.
  • Retatrutide in the U.S. (2026): University of Iowa scientists testing for CBS News found two online samples matched a reference standard, while a vial bought from a convenience store contained no peptide and grew bacteria.
  • Knockoff tirzepatide (2024): Eli Lilly reported that almost half of the knockoff tirzepatide it tested was below labeled potency, with some samples showing bacteria, high endotoxin or impurities. Lilly is not a neutral party, and it did not publish sample counts.

We found no large, peer-reviewed purity study of other popular research peptides such as BPC-157 or TB-500. That gap is part of the story.

Compounding is not a perfect benchmark

To be fair to the other side: compounding has its own documented problems. As of May 2026 the FDA had received about 990 adverse event reports involving compounded semaglutide and more than 730 involving compounded tirzepatide, and it has linked some of them to dosing errors by patients and by clinicians. The FDA notes these numbers likely undercount the problem. Small studies and lawsuits have found impurities and low potency in some compounded GLP-1 products, and, as noted above, the API sites that serve only compounders draw a large share of FDA enforcement. Oversight lowers the risk; it does not remove it.

The bottom line

  • Often the same upstream: both channels can draw API from the same countries and sometimes the same manufacturers.
  • Not the same downstream: what differs is verification of the source, testing of each lot, sterile finishing, a licensed professional and someone accountable when something goes wrong.
  • "Typically the same" is not something the evidence can show. No one has compared lots across the two channels, and the independent tests that exist show wide swings in content and contamination in grey-market products.

For anyone working with research material in the lab, the practical takeaway is the same as always: match the lot number to an independent certificate of analysis, and treat a vial with no verifiable testing as unknown material. For the regulatory background, see what the July 2026 FDA vote means.

Disclosure: Peptide Knowledge is affiliated with IronCell Research, a supplier of research compounds, and benefits when readers buy from it. Read our disclosure.

Key references

This post is educational. It does not recommend any source and does not encourage using research compounds in people or animals.

Educational content only

Peptide Knowledge is an educational reference. Nothing on this site is medical advice, a treatment recommendation or dosing guidance. Many compounds described here are not approved for human use and are sold for laboratory research only. Talk to a licensed clinician before starting, stopping or changing any medication or supplement.

All blog posts