Semaglutide vs Tirzepatide vs Retatrutide: What Is Actually Different

Retatrutide comes up in almost every peptide conversation right now, usually from someone who does not realise it is not a thing you can legally buy. That is the part that gets lost in the comparison, so it is worth saying early.

The other two you can get from a doctor. The third one you cannot get at all.

The actual difference

Each of these activates more of your body's appetite and metabolism receptors than the last.

Semaglutide hits one, GLP-1. Tirzepatide hits two, GLP-1 and GIP. Retatrutide hits three, adding glucagon.

More receptors has meant more weight loss in trials, and more side effects along with it. Everything below is a variation on that.

Semaglutide

A lab-made copy of GLP-1, a gut hormone your body releases after eating. It turns up your fullness signal so you eat less without fighting yourself the whole time. This is the drug in Ozempic and Wegovy. FDA-approved, weekly, and the most established of the three by a wide margin.

Tirzepatide

Same idea, plus a second gut hormone receptor called GIP. Two signals instead of one, and the trials showed a meaningful jump in weight loss because of it. This is the drug in Mounjaro and Zepbound. Also FDA-approved and weekly.

Retatrutide

The third receptor is the interesting one. GLP-1 and GIP mostly reduce how much you eat. Glucagon activation also pushes your body to burn more energy, so retatrutide is working on intake and expenditure at the same time rather than appetite alone.

That is the theory, and the trial numbers back it up. Eli Lilly published Phase 3 results in May 2026 from a trial called TRIUMPH-1, with 2,339 participants over 80 weeks. Average weight loss came in around 17.6% at 4 mg, 23.7% at 9 mg, and 25% at 12 mg, against 3.9% for placebo. A subgroup who kept going to 104 weeks reached roughly 30%. An earlier trial in people who also had knee osteoarthritis reported about 28.7% at 68 weeks.

Those are the largest numbers published for anything in this class.

It is also not approved anywhere in the world, which I will come back to.

Putting the numbers side by side

Two caveats, because this table is more slippery than it looks.

Nobody has published a trial putting all three head to head. These are separate studies with different populations and different lengths, so the ranking is directionally right but the gaps between the numbers are not precise.

And these are averages, from people who were also getting diet and exercise support as part of the trial. Individual results ranged enormously in every one of these studies.

The part that actually decides this

Semaglutide and tirzepatide are approved medicines. A doctor can prescribe them, a pharmacy makes them, and the vial contains what the label says.

Retatrutide has no legal route. Eli Lilly is expected to file for approval, but right now it exists legally only inside clinical trials.

So anything sold as retatrutide is a research chemical from a vendor, and every impressive number above was produced with Eli Lilly's molecule under clinical supervision. Those results say nothing about what is in a vial you bought from a website. Different molecule, potentially. Different purity, probably. Nobody checked on your behalf.

If you are going that route anyway, read how to vet a vendor first. Independent testing and a batch COA are the only things between you and an unlabelled powder.

Side effects

All three share the same core problem, which is your stomach. Nausea, vomiting, diarrhea, constipation, worst in the first weeks and again every time the dose steps up. This is why all of them start low and climb slowly over months.

The two approved ones carry a boxed warning about a rare thyroid tumour risk seen in rodents.

Retatrutide's trials also flagged dysesthesia, meaning strange skin sensations, plus urinary tract infections and a mild temporary rise in heart rate. Dropouts from side effects went from about 4% at the lowest dose to about 11% at the highest.

That climb is worth sitting with. The bigger numbers come with a real tolerance cost, and roughly one in nine people at the top dose stopped because of it.

What the trials measured, and what people actually report

Clinical trials count what they set out to count. They are much better at capturing weight and nausea than they are at capturing what it feels like to be on one of these for a year.

In 2026, researchers at the University of Pennsylvania went at this from the other direction. They analysed 410,198 Reddit posts about semaglutide and tirzepatide going back to 2019, and found 67,008 people who said they were using one of them. Just under half described at least one side effect.

What those people reported, in order:

Nausea leading is no surprise. Fatigue sitting second is, and it is the reason this data is worth looking at. Tiredness is not something the trial write-ups lead with, and it is not what the marketing prepares you for, but it is the second most common thing real users bring up unprompted. If you go in expecting stomach trouble and instead get flattened for a few weeks, it helps to know that is common rather than a sign something is wrong.

Read that table with one caveat in mind. People post about problems more than they post about nothing happening, so these percentages are not incidence rates and should not be read as "36.9% of users get nausea." They are a ranking of what people talk about, which is a different and still useful thing.

The other thing the trials never captured is the effect people describe most consistently: appetite going quiet. Not willpower improving, not being too nauseated to eat, but the constant background chatter about food simply stopping. People describe forgetting to eat, getting full halfway through a normal plate, or losing interest in a snack they used to think about daily. It is the single most repeated subjective experience in these communities, and it has no line item in any trial table.

What it costs

Sticker price tells you almost nothing here, because the dosing differs so much between them and changes as you escalate. A vial that lasts eight weeks at a starting dose might last three at a maintenance dose.

That is the part people get wrong when they budget. They price the vial once, at the dose they are starting on, and then get a surprise four months later when they are running through the same vial in half the time. Your monthly cost on these does not stay still. It climbs with your dose and then settles wherever you land.

Cost per dose and cost per month are the only comparisons worth making. Put your vial size, dose, and price into the calculator and it will work out what you are actually spending and when you run out. Then change the dose to where you expect to end up and run it again, because that second number is the one you will be living with.

Where that leaves you

Going from one receptor to two to three has produced bigger weight loss each time, with more side effects each time. That pattern is real and well evidenced.

But two of these are medicines and one is not, and that gap matters more than the percentages. A verified 15% beats an unverified 28% every time.