One receptor, two receptors, three receptors. This is the highest-intent comparison in the entire incretin category, and most versions of it online quietly skip the two facts that matter most: the three drugs sit at very different regulatory stages, and their headline weight-loss numbers come from different trials that cannot be directly stacked against each other. This page compares mechanism, evidence, safety, and availability honestly — including what we do not yet know.
Every row below is drawn from the full compound profiles on this site — FDA prescribing information for the two approved drugs, and published TRIUMPH Phase III data for retatrutide. Weight-loss figures are each compound's own pivotal-trial result, not head-to-head data; no completed trial has compared all three directly.
Semaglutide Novo Nordisk FDA-Approved |
Tirzepatide Eli Lilly FDA-Approved |
Retatrutide Eli Lilly Investigational |
|
|---|---|---|---|
| Class / Mechanism | GLP-1 receptor agonist (single receptor) | Dual GIP/GLP-1 receptor co-agonist (unimolecular) | Triple GIP/GLP-1/glucagon receptor agonist (unimolecular) |
| FDA Regulatory Status | Approved: Ozempic® (2017, T2D), Rybelsus® (2019, T2D), Wegovy® (2021, weight; 2024 CV-risk expansion) | Approved: Mounjaro® (2022, T2D), Zepbound® (2023, weight; Dec 2024 OSA expansion) | Not approved in any form. Phase III complete; Lilly plans a BLA submission in Q1 2027 |
| Brand Names | Ozempic® · Wegovy® · Rybelsus® | Mounjaro® · Zepbound® | None — development code LY3437943 only |
| Evidence Level | Level I — FDA-approved on multiple independent Phase III RCTs (STEP, SUSTAIN, PIONEER, SELECT) | Level I — FDA-approved on the SURPASS and SURMOUNT Phase III programs | Level Ib — strong Phase III RCTs (TRIUMPH-1 to -4) with no completed regulatory review |
| Key Trials | STEP 1 (n=1,961) · SELECT (n=17,604) · SUSTAIN-6 · PIONEER 6 | SURMOUNT-1 (n=2,539) · SURPASS-2 (head-to-head vs. semaglutide 1 mg dose tier) · SURMOUNT-OSA | TRIUMPH-1 (n=2,339) · TRIUMPH-2 (T2D) · TRIUMPH-3 (CVD) · TRIUMPH-4 (knee OA) |
| Route | Once-weekly subcutaneous (Ozempic®/Wegovy®) or once-daily oral tablet (Rybelsus®) — approved label | Once-weekly subcutaneous — approved label | Once-weekly subcutaneous — clinical-trial protocol only; no approved route exists |
| Pivotal Weight-Loss Result | −14.9% body weight vs. −2.4% placebo (STEP 1, 68 weeks) | −20.9% vs. −3.1% placebo at highest approved dose (SURMOUNT-1, 72 weeks) | −28.3% at 80 weeks; −30.3% at the 104-week extension, highest trial arm (TRIUMPH-1 — trial data, not an approved result) |
| Common Side Effects | Nausea (~44% in STEP 1), vomiting, diarrhea — mostly mild-to-moderate, titration-phase | Nausea (~31%), diarrhea (~23%) in SURMOUNT-1 — generally mild-to-moderate | Nausea 42.4%, diarrhea 32.0%, vomiting 25.3% at the highest TRIUMPH-1 arm — highest GI burden of the three |
All three compounds are engineered long-acting peptides built on the same pharmacokinetic playbook — a DPP-4-resistant backbone plus a fatty-diacid side chain that binds serum albumin, stretching a minutes-long hormone signal into a once-weekly drug. What separates them is how many receptors of the gut-hormone (incretin) axis each one engages, and each added receptor has, so far, meant a larger trial effect size and a heavier tolerability cost.
Semaglutide is a modified GLP-1 analogue, 94% identical to the native hormone, acting on one receptor expressed across four organ systems. In the pancreas it amplifies glucose-dependent insulin secretion and suppresses glucagon — glucose-dependent, which is why hypoglycemia risk is low as monotherapy. In the stomach it delays gastric emptying (an effect that attenuates over weeks). In the hypothalamus and brainstem it activates satiety circuitry — the primary driver of sustained weight loss. And in cardiovascular tissue, GLP-1R engagement is the proposed basis for the 20% relative reduction in major adverse cardiovascular events demonstrated in the SELECT trial. That last point matters for this comparison: semaglutide is the only compound of the three with a completed hard cardiovascular-outcomes trial.
Tirzepatide is a single 39-amino-acid peptide built primarily on the native GIP backbone, engineered to also activate the GLP-1 receptor — a genuine unimolecular dual agonist, not two drugs combined. Its GLP-1R potency is deliberately calibrated lower than semaglutide's, on the theory that the GIP arm compensates and the balanced dual signal buys a better efficacy-to-tolerability trade. GIP receptors are dense on adipocytes, where agonism appears to improve adipose insulin sensitivity and healthy lipid storage (rather than lipid spilling into liver and muscle), and are also present in hypothalamic appetite regions, where they may contribute an independent satiety effect. The clinical result — a roughly 21% average weight reduction in SURMOUNT-1 versus roughly 15% in semaglutide's STEP 1 — is the strongest real-world validation to date that adding an incretin receptor adds efficacy.
Retatrutide extends the same design logic one receptor further, adding glucagon receptor (GCGR) agonism to the dual-incretin foundation. This is the conceptually risky piece: glucagon is a catabolic, glucose-raising hormone, and pairing it with two glucose-lowering signals is a deliberately engineered balancing act. The intended payoff is hepatic — glucagon agonism drives fatty-acid oxidation in the liver, producing meaningful liver-fat reductions in trial substudies — plus a modest increase in resting energy expenditure. The bet is that the GIP/GLP-1 arms absorb glucagon's glycemic downside while its thermogenic and lipolytic effects are kept. TRIUMPH-1's effect size suggests the bet is paying off on efficacy; the elevated GI adverse-event and discontinuation rates show the tolerability bill that came with it. No head-to-head trial against tirzepatide has been completed, so the size of the true incremental benefit remains a cross-trial inference.
The three pivotal obesity trials are the numbers everyone quotes, so here they are on one scale — with the caveat that belongs in the same breath: these are three different trials, run in different years, with different durations, populations, and highest-dose arms. Cross-trial comparison is a reasonable hypothesis-generator and a poor superiority claim.
Semaglutide has the deepest evidence base of the three by a wide margin. Beyond STEP 1's −14.9% at 68 weeks (with just over half of participants losing at least 15%), the SELECT trial (n=17,604) demonstrated a 20% relative reduction in major adverse cardiovascular events in non-diabetic adults with established cardiovascular disease — the first cardiovascular-outcomes benefit shown for this class outside type 2 diabetes, and the basis for Wegovy®'s 2024 label expansion. SUSTAIN-6 and PIONEER 6 anchor the diabetes and oral-formulation evidence respectively.
Tirzepatide delivered −20.9% at the highest dose arm in SURMOUNT-1, with roughly 57% of that arm losing at least 20% of body weight. It also holds the only completed head-to-head trial among these three — SURPASS-2, which beat semaglutide on HbA1c and weight in type 2 diabetes. The honest caveat from our Tirzepatide profile applies: SURPASS-2 compared tirzepatide against semaglutide's lower 1 mg diabetes dose tier, not the higher weight-management dose, so it does not settle the weight-loss question at full doses. SURMOUNT-OSA additionally made tirzepatide the first drug ever approved for obstructive sleep apnea.
Retatrutide posted the largest numbers of any incretin-class agent to date: −28.3% at 80 weeks in TRIUMPH-1's highest arm, deepening to −30.3% (an average of 85 pounds) in the 104-week extension — figures that approach bariatric-surgery territory. TRIUMPH-2 extended efficacy to type 2 diabetes (−20.8%, HbA1c reductions up to 1.6 points), and TRIUMPH-3 showed strong cardiometabolic risk-factor movement (triglycerides −37%, systolic blood pressure −9.3 mmHg) in patients with established cardiovascular disease. But these are risk-factor and weight endpoints, not outcomes: retatrutide has no SELECT-style completed cardiovascular-outcomes trial, no completed FDA review, and no safety data beyond 104 weeks. For clarity on what was actually studied: TRIUMPH-1's arms were 4 mg, 9 mg, and 12 mg weekly — clinical-trial cohort doses reached through investigator-supervised escalation inside the trial, cited here strictly as trial-design fact, not as anything available or appropriate outside a registered trial.
Patients tend to frame this as "which drug is strongest?" Physicians frame it differently: which approved drug fits this patient's comorbidities, risk profile, insurance reality, and tolerance for GI side effects — and is there any reason to look past the two approved options at all? Here is the shape of that reasoning. None of it substitutes for an individual consultation.
Two threads run through every version of this conversation. First, insurance coverage and supply frequently decide between the two approved drugs before pharmacology does — a practical reality worth raising early with the prescriber. Second, more efficacy has consistently meant more GI burden across this class, so the "strongest" agent is not automatically the right one for a patient who values tolerability, or who needs to keep lean mass and nutrition intact during rapid loss. Every pathway here ends in the same place: talk to a licensed physician.
The three compounds share one safety architecture — gastrointestinal side effects concentrated in dose-escalation phases, low intrinsic hypoglycemia risk as monotherapy, and the same rodent-derived thyroid C-cell tumor signal across the class. The differences are of degree, and they track receptor count.
| Trial-Reported Effect | Semaglutide (STEP 1) | Tirzepatide (SURMOUNT-1) | Retatrutide (TRIUMPH-1, highest arm) |
|---|---|---|---|
| Nausea | ~44% | ~31% | 42.4% (vs. 14.8% placebo) |
| Diarrhea | 15–30% (with vomiting) | ~23% | 32.0% (vs. 13.5% placebo) |
| Vomiting | 15–30% (with diarrhea) | Reported, dose-dependent | 25.3% (vs. 4.8% placebo) |
| Discontinuation for adverse events | Low, titration-managed | Low, titration-managed | 11.3% (vs. 4.9% placebo) — highest of the three |
When patients bring me this comparison, they almost always arrive holding the three headline numbers — 15, 21, 30 — as if the choice were already made. My first job is usually to complicate that a little. Those numbers come from three different trials, and the drug with the biggest one cannot be prescribed at all. What I actually see in this progression is the cleanest natural experiment in modern metabolic pharmacology: each added receptor has bought more efficacy and cost more tolerability, on an almost dose-response-like curve. That is scientifically beautiful, and it is also exactly why I resist ranking these drugs on a single axis.
Between the two approved agents, I find the honest answer is that both are excellent and the tiebreakers are usually not pharmacological — they are the patient's cardiovascular history, their sleep apnea, their insurance formulary, their feelings about needles, and how their gut handles the first months. Semaglutide's SELECT data gives it an outcomes pedigree nothing else in the class can match yet; tirzepatide's efficacy at approved doses is the strongest available. On retatrutide, my position is the same one I hold across this site: I am genuinely impressed by the TRIUMPH data, and impressed is not the same as available. A 30% average reduction with an 11% adverse-event discontinuation rate and no data past two years is a compound worth watching closely from inside the regulatory process — not one worth improvising with from outside it. If this class is on your mind, take this page to a licensed physician and have the version of this conversation that includes your bloodwork.