Six molecules with a current United States label act on the GLP-1 receptor. That is the whole answer to "which drugs are GLP-1s", and it is short. What makes the question worth a page is that the popular list is longer than the real one in one direction and shorter in another: a drug most people would name first is not only a GLP-1, and a drug many people assume belongs here does not act on the receptor at all. The list of drugs still in trials is longer and changes by acquisition as well as by approval: Pfizer's candidate came with its purchase of Metsera.
This page is a register, read from the FDA's DailyMed label database on 2026-09-16 and dated accordingly. It counts molecules, not brand names, and it says how it counted.
Counting molecules is only half the disambiguation problem. The abbreviation itself carries four separate meanings — a hormone, a receptor, this drug class and a marketing label on products containing none of it — and since this register was compiled, a brand name has stopped identifying even the dosage form.
The register, as read on 2026-09-16
| Molecule | Current US brands | What its own label calls it | Distinct labelers | Generic |
|---|---|---|---|---|
| Semaglutide | Ozempic, Wegovy, Rybelsus | GLP-1 analogue, GLP-1 receptor agonist | 2 | no |
| Tirzepatide | Mounjaro, Zepbound | GIP receptor and GLP-1 receptor agonist | 2 | no |
| Dulaglutide | Trulicity | GLP-1 receptor agonist | 2 | no |
| Liraglutide | Victoza, Saxenda, Xultophy | GLP-1 receptor agonist | 13 | yes |
| Exenatide | Byetta | GLP-1 receptor agonist | 3 | yes (one) |
| Lixisenatide | Soliqua (combination only) | GLP-1 receptor agonist | 1 | no |
Metformin is not in the table, and the reason is below.
The three answers
Ask "is this drug a GLP-1" of the medicines people actually ask about and three different answers come back.
Yes, straightforwardly. Semaglutide, dulaglutide, liraglutide, exenatide and lixisenatide each bind one receptor, and it is the GLP-1 receptor. Ozempic, Wegovy and Rybelsus are all semaglutide; Victoza and Saxenda are both liraglutide (Saxenda's label, trials and generics are read separately). The brand count is larger than the molecule count because the same molecule is approved separately for different indications, which is a regulatory fact rather than a pharmacological one. Dulaglutide and semaglutide are also the one pair in this list with a completed head-to-head trial, set out on Trulicity vs Ozempic. Dulaglutide's label, pens and heart trial are read in full on what Trulicity is. The dates on which each of those brands, and each later indication, was approved are tabled in the GLP-1 approval timeline.
Not only. Tirzepatide's label opens its Mechanism of Action section with a sentence that settles the question and is quoted here in full: "Tirzepatide is a GIP receptor and GLP-1 receptor agonist." It goes on to say that tirzepatide "selectively binds to and activates both the GIP and GLP-1 receptors, the targets for native GIP and GLP-1". Mounjaro and Zepbound are that molecule. Whether activating GIP is what gives tirzepatide its edge is now disputed by a drug that blocks the same receptor; see GIP and GLP-1. Describing them as GLP-1s is the shorthand everyone uses, including pharmacists, and it is not exactly wrong — but it is the kind of shorthand that becomes wrong the moment a result from a single-receptor drug is carried across to it, or the reverse. What activating each receptor does in the pancreas, the stomach and the brain is covered in our mechanism explainer.
No. Metformin does not act on the GLP-1 receptor.
Metformin, which is the one people ask about
Metformin turns up in this question more than any other non-GLP-1, and the reason is circumstantial: it is the old, cheap, widely prescribed type 2 diabetes medicine that many people were taking before these drugs existed, so it sits in the same conversation. It does not sit in the same class.
Its own label is explicit about the difference. Metformin decreases hepatic glucose production, decreases intestinal absorption of glucose, and improves insulin sensitivity — and the label states that its pharmacologic mechanisms of action "are different from other classes of oral antihyperglycemic agents". There is no GLP-1 receptor anywhere in that description, because metformin does not go near it. Its modest effect on weight is a separate question with its own trial record, set out on does metformin cause weight loss.
What is generic today, and what is not
The generic question has a cleaner answer than the general debate suggests, and the answer is dated.
On 2026-09-16, liraglutide carried 22 current DailyMed listings from 13 distinct labelers, and eleven of those labelers are companies other than Novo Nordisk — Teva, Biocon, Cipla, Hikma, Lupin, Meitheal, NorthstarRx, Orbicular and Hybio among them. That is not a single generic entrant; it is a crowded generic field, and it has arrived on the oldest daily molecule in the class rather than on the ones people are waiting for.
Exenatide has one generic label, from Amneal, alongside AstraZeneca's Byetta — but a label on DailyMed is not proof of sale: on 14 October 2026 FDA's Drugs@FDA listed both, and every other exenatide product, as discontinued (the exenatide record). It is also the oldest molecule here and the only one copied from an animal peptide — exendin-4, first isolated from Gila monster venom in 1992.
Semaglutide, tirzepatide and dulaglutide have none. Each returned listings from exactly two labelers, and in every one of the three cases the second is a repackager rather than a second manufacturer. The patent timeline behind that is a separate question and we treat it separately on when Ozempic goes generic.
Two further things the register shows and the general conversation does not. Lixisenatide has no standalone listing at all: its single current listing is Sanofi's SOLIQUA 100/33, where it is combined with insulin glargine, so in the United States it reaches patients only inside an insulin product — one of two combination pens that are partly insulin and partly GLP-1. And exenatide's once-weekly extended-release form returned no current listing either — the Amneal generic is the twice-daily form, as its own label's dosing section makes clear, not the weekly one. Both statements describe what the database held on the date read, which is the only thing a register can honestly claim; neither is a statement about why.
How this was counted, and one trap in the counting
Every figure above comes from the DailyMed SPL service, queried by generic drug name on 2026-09-16, with the full result set read rather than the first page.
The trap is worth naming because it would have changed several numbers. A DailyMed listing is not a product. Repackagers file their own labels for medicines they did not manufacture, and one repackager — A-S Medication Solutions — accounts for six of semaglutide's nine listings, six of tirzepatide's eight, and six of dulaglutide's seven. Counted naively, semaglutide looks like a nine-label molecule with a busy field behind it. Counted by distinct labeler, it is one manufacturer and one repackager. The table above reports distinct labelers for that reason, and liraglutide's 13 is the figure that survives the correction rather than one produced by it.
For what these molecules have in common once the classification is settled — a hormone the body already makes, and the very short life it has in circulation — see natural GLP-1. For the drugs sold as GLP-1s without being any of the six, see GLP-1 supplements and GLP-1 patches.
Being one of the six is also not the same as being obtainable. Each label names who the product is for and, in a boxed warning, who it is forbidden to, and those two sentences do more gatekeeping than price does — the four gates between a person and a prescription reads them out.
Nothing on this page is a dose, a schedule or a recommendation, and nothing here says which of these medicines is appropriate for anyone. Corrections go to the contact page.