Comparison
Mazdutide vs. Retatrutide
Mazdutide or retatrutide – which works better, and which is approved?
The short answer
Mazdutide acts on two receptors, GLP-1 and glucagon; retatrutide additionally acts on GIP. Both originally come from the Eli Lilly lab. Mazdutide has been approved in China since 2025 and reached 16.65 percent weight loss after 60 weeks in the phase 3 trial GLORY-2, versus 1.50 percent on placebo. Retatrutide reached 25.0 percent versus 3.9 percent after 80 weeks in TRIUMPH-1, but is not approved anywhere. The gap is partly misleading: in the two most comparable Lilly trials, each phase 2 over 48 weeks, the highest tested levels landed at a similar level, 22.3 and 24.2 percent. There is no trial comparing the two directly. Neither is a marketable medicine in Germany, and grey-market products are not the study drug.
The numbers side by side
| Axis | Mazdutide | Retatrutide | Difference |
|---|---|---|---|
| Human evidenceHow much and how good the data in humans are | 8 | 9 | Retatrutide better studied |
| MechanismHow well the chain of action is understood | 8 | 8 | even |
| Safety dataHow well safety has been studied – not how safe it is | 6 | 7 | Retatrutide better studied |
| Hype gap10 means: the advertising stays level with the data | 4 | 4 | even |
| Track record of useHow much documented use in humans exists – does not measure whether it works | 4 | 4 | even |
| Rating | Well supported, heavily overhyped · Well supported, heavily overhyped | ||
“Track record of use” measures how long and how widely something has been used in humans – not whether it works. The BK-Score is a subjective assessment by Biohacking Kompakt based on published scoring rules (German). It is not a scientific rating, not the consensus of a professional society and not a medical recommendation.
How the two work
Both belong to a new generation of weight-loss drugs that combine the satiety hormone GLP-1 with glucagon. Mazdutide is derived from oxyntomodulin, a gut hormone that naturally acts on both the GLP-1 and the glucagon receptor. Retatrutide is a triple agonist at GLP-1, GIP and glucagon, in other words the principle of tirzepatide plus glucagon.
GLP-1 curbs appetite and slows gastric emptying, GIP strengthens the insulin response, and the glucagon arm is meant to mobilize energy from the liver and break down fat there. How much the glucagon part contributes on its own has not been tested separately for either drug, and nobody has yet measured higher energy expenditure in humans for mazdutide.
The two share an origin: mazdutide was designed at Lilly and licensed to Innovent Biologics for China in 2019, where the entire approval programme took place. According to trade press, Lilly itself is putting more research resources into retatrutide.
The best-matched pair: two Lilly trials over 48 weeks
First of all: there is no trial that compares mazdutide and retatrutide directly, neither completed nor registered on ClinicalTrials.gov. Everything on this page is placed side by side. The mazdutide phase 3 trials were run almost entirely in China, with a baseline weight of about 87 to 94 kilos; those on retatrutide were international and ran for 80 instead of 24 to 60 weeks, with heavier participants and partly different analysis rules.
The most comparable are two phase 2 trials that Lilly ran itself: both focused on the US, both over 48 weeks, both in adults without diabetes with a BMI of 30 or more, or 27 or more with a comorbidity.
| Feature | Mazdutide (US phase 2) | Retatrutide (phase 2, NEJM 2023) |
|---|---|---|
| Participants | 179 | 338 |
| Weight after 48 weeks | −10.5 / −19.2 / −22.3 % | −8.7 / −17.1 / −22.8 / −24.2 % |
| Placebo | −0.05 % | −2.1 % |
| Highest level minus placeboour own calculation, not a trial figure | about 22.3 percentage points | about 22.1 percentage points |
| Source of the value | Results in the trial registry, secondary endpoint, only those who continued treatment as planned | Journal article; analysis rule not stated in the abstract |
In this pair, the highest levels land at a similar level, each about 22 percentage points above placebo. The large differences in the headlines, roughly 14 versus 25 percent, come mainly from the fact that the Chinese mazdutide trials used lower levels, shorter durations and lighter participants. That is a pointer, not proof: the mazdutide value after 48 weeks is a secondary endpoint from a small trial with 51 people at the highest level, calculated across everyone who continued treatment as planned, and at that level one in five stopped treatment because of side effects. At the primary endpoint, week 32, the three mazdutide levels were at minus 7.3, 15.6 and 18.1 percent versus minus 0.9 percent.
A network meta-analysis in BMJ Medicine 2026 puts mazdutide at 13.16 and retatrutide at 22.10 percent weight loss versus placebo. For mazdutide, however, only a lower level from China was included; the US trial appeared after the search date. Neither is enough for a ranking.
Phase 3: where each drug stands
Mazdutide. Four phase 3 trials have been published after peer review, all from China. GLORY-1 (NEJM 2025) with 610 participants: after 48 weeks, minus 11.00 and minus 14.01 percent versus plus 0.30 percent. GLORY-2 (JAMA 2026) with 461 people treated with a BMI of 30 or more: after 60 weeks, minus 16.65 versus minus 1.50 percent. In type 2 diabetes, mazdutide lowered HbA1c by 1.57 and 2.15 percentage points in DREAMS-1, versus 0.14 on placebo, and did better than dulaglutide in DREAMS-2 (both Nature 2026). Against semaglutide at the diabetes dose, it was superior in DREAMS-3 according to the manufacturer: 48.0 versus 21.0 percent reached the combined goal of HbA1c below 7 percent and at least 10 percent weight loss, and weight fell by 10.29 versus 6.00 percent. The trial was open-label; so far only the study protocol has been peer-reviewed. Ongoing are GLORY-3 against semaglutide in overweight with fatty liver (primary completion end of 2026) and phase 3 trials on sleep apnea, in adolescents and on high blood pressure. Lilly has not registered any phase 3 trial outside China.
Retatrutide. Peer-reviewed are TRIUMPH-1 (NEJM 2026) with 2,339 participants without diabetes: after 80 weeks, minus 17.6, 23.7 and 25.0 percent versus minus 3.9 percent, calculated across all randomized participants. Also TRIUMPH-2 in type 2 diabetes (Lancet 2026, minus 18.8 versus minus 5.1 percent), the diabetes trial TRANSCEND-T2D-1 (Lancet 2026) and the sleep apnea analysis (NEJM 2026). Known only from manufacturer announcements are TRIUMPH-4 on knee osteoarthritis and the first head-to-head comparison, TRANSCEND-T2D-2, against semaglutide at the highest Ozempic dose: minus 19.5 and 20.9 versus minus 9.4 percent after 80 weeks. The comparison with tirzepatide, TRIUMPH-5, is still pending; more on this in the comparison of retatrutide and tirzepatide.
Liver fat. Here both have measured data from magnetic resonance imaging. On mazdutide, liver fat in the US trial fell by 71.2 to 80.2 percent after 48 weeks in participants with at least 10 percent liver fat, versus 26.3 percent on placebo. On retatrutide, it fell by 82.4 percent at the highest level after 24 weeks in a phase 2 substudy, while it rose by 0.3 percent on placebo. There are no data on heart attack, stroke or mortality for either drug.
Side effects and discontinuations
For both, nausea, vomiting and diarrhea are the main issue. On mazdutide in GLORY-2, 53.1 percent had vomiting, 46.9 percent nausea and 39.4 percent diarrhea, mostly mild to moderate. 2.9 percent stopped because of side effects there, and 0.5 and 1.5 percent in GLORY-1 versus 1.0 percent on placebo. At the highest level of the US trial, by contrast, 20 percent stopped treatment because of side effects. In a Chinese phase 2 trial, the pulse rose by 5.38 to 8.75 beats per minute after 24 weeks, versus 4.81 on placebo.
For retatrutide, the manufacturer reports nausea in 42.4 percent at the highest level of TRIUMPH-1 and discontinuations due to side effects of up to 18.2 percent in TRIUMPH-4. A special feature is dysaesthesia, an abnormal skin sensation, in 4 to 12.5 percent depending on the trial. There is no peer-reviewed phase 3 figure for heart rate yet.
The network meta-analysis found no increased discontinuation rate for mazdutide and an increased one for retatrutide. The authors, however, regard this endpoint as exploratory only because of inconsistencies in the network, and the trials define discontinuation differently: treatment stopped, trial left, or both. A fair comparison of the rates is therefore not possible.
Approval and grey market
Mazdutide has been approved in China for weight management since June 2025, under the name Xinermei, and since September 2025 also for type 2 diabetes; an application for a higher level has been under review there since November 2025. It is not approved anywhere else. Retatrutide is not approved anywhere in the world; Eli Lilly plans to file with the FDA in the first quarter of 2027. In Germany, neither is a marketable medicine.
The US Food and Drug Administration explicitly names both among the substances that are illegally labelled “for research purposes” or “not for human consumption” and still sold to consumers with instructions for use (FDA page as of 1 October 2026). Such products are not the study drug, and nobody controls their content or purity. For retatrutide there is a peer-reviewed case report and US media reports on laboratory tests; for mazdutide there is no published analysis of grey-market products so far.
Why these numbers
Mazdutide
Four published phase 3 trials met their primary endpoints: GLORY-1 (NEJM 2025) randomized 610 Chinese adults over 48 weeks, with 11.00 to 14.01 percent weight reduction versus plus 0.30 percent on placebo; GLORY-2 (JAMA 2026) showed 16.65 percent after 60 weeks with a higher dose, versus minus 1.50 percent. In type 2 diabetes, mazdutide lowered HbA1c versus placebo (DREAMS-1, 320) and more than dulaglutide (DREAMS-2, 731); against semaglutide at the diabetes dose it did better in DREAMS-3 according to the manufacturer, though so far only the study protocol has been peer-reviewed. In China, the substance is approved for obesity and type 2 diabetes. Dual GLP-1 and glucagon receptor agonism has been confirmed in humans, and in the US phase 2 trial liver fat measured by MRI fell by up to 80.2 percent; higher energy expenditure has not been measured. The phase 3 trials were conducted exclusively in China, and outside China there is one US phase 2 trial with 179 participants; in Germany there is no approval – a point sales pages regularly leave out.
Retatrutide
The strongest weight data ever collected, now also peer-reviewed in phase 3. Phase 2: 338 participants, 48 weeks, 24.2 percent versus 2.1 percent on placebo (NEJM 2023). TRIUMPH-1 (NEJM 2026): 2,339 participants without diabetes, 80 weeks, double-blind, 25.0 percent at the highest dose versus 3.9 percent on placebo, calculated across all randomized participants; according to the manufacturer, 28.3 percent among everyone who continued treatment as planned. Also peer-reviewed are TRIUMPH-2 in type 2 diabetes (Lancet 2026, 18.8 versus 5.1 percent) and the diabetes trial TRANSCEND-T2D-1 (Lancet 2026). According to the manufacturer, TRIUMPH-4 reached 28.7 percent versus 2.1 percent on placebo with 445 participants over 68 weeks. Triple agonism is demonstrated in humans; the contribution of the glucagon arm has not been tested separately. The price is in the same trials: nausea in up to 43.2 percent (highest dose in TRIUMPH-4) and discontinuation due to side effects in up to 18.2 percent. Safety 7 (previously 6): these data now come from several peer-reviewed phase 3 trials, TRIUMPH-1 alone with 2,339 participants over 80 weeks; for a higher score, long-term data from approval and pharmacovigilance are missing. There is no approval yet; what is traded on the grey market today is not the study drug.
Frequently asked questions
Is mazdutide weaker than retatrutide?
Going by the headlines, yes: 16.65 percent in GLORY-2 versus 25.0 percent in TRIUMPH-1. But these trials differ in duration, levels and participants. In the two most comparable phase 2 trials by Lilly, the highest levels after 48 weeks landed at a similar level, 22.3 percent for mazdutide and 24.2 percent for retatrutide, each about 22 percentage points above placebo. A head-to-head comparison is missing, and no ranking can be derived from this.
Is there a trial that compares mazdutide and retatrutide directly?
No. No trial with both substances is registered on ClinicalTrials.gov, and the literature contains only reviews and meta-analyses. Direct comparisons exist only with other drugs: mazdutide against dulaglutide and, according to the manufacturer, against semaglutide; retatrutide, according to the manufacturer, against semaglutide. Results against tirzepatide are still pending for both.
Is mazdutide or retatrutide approved in Germany?
No, neither. Mazdutide has so far been approved only in China, for weight management since June 2025 and for type 2 diabetes since September 2025. Retatrutide is not approved anywhere in the world; Eli Lilly plans to file with the FDA in the first quarter of 2027. What is offered online in Germany is untested goods.
Which of the two has more side effects?
For both, nausea, vomiting and diarrhea are the main issue. Mazdutide stands out in GLORY-2 for frequent vomiting, 53.1 percent, with few discontinuations at 2.9 percent; at the highest level of the US trial, however, 20 percent stopped treatment because of side effects. For retatrutide, the manufacturer reports nausea in 42.4 percent and discontinuations of up to 18.2 percent, plus abnormal skin sensations. Because trials, levels and definitions differ, the rates cannot be compared directly.
Sources
- Ji et al., N Engl J Med 2025 (GLORY-1)
- Gao et al., JAMA 2026 (GLORY-2)
- Zhu et al., Nature 2026 (DREAMS-1)
- Guo et al., Nature 2026 (DREAMS-2)
- Luo et al., Contemp Clin Trials 2025 (DREAMS-3, study protocol)
- Innovent, manufacturer announcement on DREAMS-3, 26 Oct 2025
- Hsia et al., Lancet Diabetes Endocrinol 2026 (US phase 2)
- ClinicalTrials.gov NCT06124807 (US phase 2 results after 48 weeks)
- Ji et al., Nat Commun 2023 (mazdutide phase 2, heart rate)
- Shirley, Drugs 2025 (Mazdutide: First Approval)
- Jastreboff et al., N Engl J Med 2023 (retatrutide phase 2)
- Jastreboff et al., N Engl J Med 2026 (TRIUMPH-1)
- Bellido et al., Lancet 2026 (TRIUMPH-2)
- Grunstein et al., N Engl J Med 2026 (sleep apnea)
- Bajaj et al., Lancet 2026 (TRANSCEND-T2D-1)
- Sanyal et al., Nat Med 2024 (retatrutide, liver substudy)
- Eli Lilly, manufacturer announcement on TRIUMPH-4, 11 Dec 2025
- Eli Lilly, manufacturer announcement on TRANSCEND-T2D-2, 28 Oct 2026
- Eli Lilly, manufacturer announcement on TRIUMPH-2 and the FDA filing, 29 Sep 2026
- Chen et al., BMJ Med 2026 (network meta-analysis)
- FDA, FDA’s Concerns with Unapproved GLP-1 Drugs Used for Weight Loss (as of 1 Oct 2026)
See all ratings in the ranking (German app)
Information only, not medical advice and not a usage or dosage recommendation. Prescription-only and unapproved substances belong in a doctor's hands.