Four with published results, and only one of them is buyable. A network meta-analysis of 58 randomized trials covering 24,214 adults put retatrutide at the top, at 22.10% of body weight against placebo (95% CI −25.60 to −18.60) [1]. Tirzepatide came second at 19.28% and CagriSema third at 17.32%, and behind them sit VK2735, UBT251 and ecnoglutide. VK2735 took off 14.7% in 13 weeks [2], UBT251 up to 13.48 kg in 12 [3], and ecnoglutide reached 13.2% at week 40 in a trial run entirely in China [4]. Of all of those, tirzepatide is the only one anyone can obtain, which makes the podium a description of the field rather than a set of choices.
This is the most complete ranking of these drugs published so far, and reading it as a shopping list produces a strange result: most of the podium is unavailable. It is a map of a menu that is still being written, and the drugs on it arrive years apart.
What was pooled
Fifty-eight randomized trials covering 24,214 adults, searched across three databases from 2000 to March 2026, all in people with overweight or obesity and without diabetes. Trials enrolling people with diabetes, or where diabetes status was unclear, were excluded, which makes the comparison cleanly about weight and means none of it describes a diabetes population [1].
Against placebo, retatrutide produced the largest reduction at 22.10%, 95% CI −25.60 to −18.60. Tirzepatide came next at 19.28%, 95% CI −20.39 to −18.16, and CagriSema, a combination of cagrilintide and semaglutide, at 17.32%, 95% CI −19.32 to −15.32. The conventional GLP-1 receptor agonists were more modest, and waist circumference and lipid outcomes followed similar patterns.
The trade at the top
Retatrutide produced the most weight loss and, on low-certainty evidence, also had higher rates of stopping treatment; danuglipron showed the same pattern, while mazdutide was better tolerated.
That is worth registering before anybody concludes the next generation is simply better. A drug that takes off more weight and that more people stop taking has not obviously beaten a gentler one, and the tolerability evidence here is explicitly graded low.
The drugs behind the podium, and how long each was tested
VK2735 pairs GLP-1 with GIP, the same combination as tirzepatide, which makes it a competitor rather than a new mechanism. In a 13-week phase 2 dose-ranging trial, weight fell 14.6 kg (14.7%) at 15 mg against 1.8 kg (1.7%) on placebo, and 130 of 140 treated participants lost at least 5% against 4 of 34 on placebo [2]. Thirteen weeks is a quarter of the 68 weeks behind every approved-drug figure on this site, and the full caution is in the VENTURE result.
UBT251 activates three receptors at once, adding glucagon to GLP-1 and GIP, which is a real mechanistic step rather than another pairing of the same two. Its first-in-human study gave weekly doses for twelve weeks and reported mean losses of 8.96 to 13.48 kg, against a 1.57 kg gain on placebo [3]. Safety was the primary endpoint and weight a secondary readout, and the abstract does not say how many people were in each dose arm, all of which is set out in the first-in-human study.
Ecnoglutide is engineered to favor one downstream signaling pathway, with the intention of separating the metabolic effects from the gastrointestinal ones. Across 664 adults at 36 Chinese centers, week-40 weight loss reached 13.2% against 0.1% on placebo [4]. Adverse events occurred in 93% of each treated group against 84% on placebo, which is broadly the profile of the existing drugs rather than an escape from it. Why its percentages are not comparable to the ones you know is in the ecnoglutide trial.
What the authors say about their own ranking
That confidence intervals overlapped for several comparisons, that head-to-head evidence is limited, and that residual uncertainty should be considered when interpreting comparative effects. Their framing is a probabilistic hierarchy rather than an order of merit.
The intervals are worth reading. Tirzepatide runs from −20.39 to −18.16, which is tight, while retatrutide runs from −25.60 to −18.60, wide enough to reach down into tirzepatide’s range. The difference between first and second place is less settled than the order suggests, which is the standing problem with any pooled ordering.
What a reader can use today
Two things, and the first is that tirzepatide’s position is solid — second place with the narrowest interval on the board, from the largest body of trials, and it is buyable. That is a stronger practical statement than retatrutide’s first place.
The modest tier is where most purchasing actually happens, because semaglutide and the older agonists are what this market sells, and their own evidence is the relevant evidence for almost everybody reading. What is available in which molecule and format is a separate question with a concrete answer.