Nobody knows yet, because it has never been measured in people. The approved semaglutide tablets absorb roughly 0.4% to 2% of the dose, and only because they carry an absorption enhancer [1]. The one published test of sublingual semaglutide used rats. There, a dose held under the tongue reached 0.29% to 0.34% of the exposure from an injection [4].
A claim that a drop works needs a measured blood level or a measured weight change in people, and neither exists for a sublingual product. Whether the drop costs less than the approved tablet is a separate question, covered in whether the GLP-1 pill is cheaper than the injection.
Why the approved tablet needs an absorber
Semaglutide is a peptide, and the gut breaks peptides down. The tablet solves this with sodium N-[8-(2-hydroxybenzoyl) aminocaprylate], known as SNAC. The label states that the tablets are co-formulated with SNAC and that absorption occurs mainly in the stomach [1]. It puts absolute bioavailability at 0.4% to 1% for Rybelsus and 1% to 2% for Ozempic tablets.
A trial in people and dogs located where that happens. Absorption took place in the stomach, close to the tablet surface, and required SNAC in the same tablet [2]. SNAC buffers the area around the tablet against digestive enzymes, and its effect on absorption is brief. Most of that study’s authors list Novo Nordisk, the maker of the tablet.
Even with SNAC, the fraction absorbed is small. A model built from six clinical pharmacology trials put it at 0.8% under the label’s dosing conditions [3]. Those conditions are a fast of 30 minutes after the dose and no more than 120 mL of water. Within-subject variability in bioavailability was 137%, a large swing from one dose to the next. By injection, the Wegovy label puts semaglutide’s bioavailability at 89% [7].
The label adds a rule that bears directly on drops. Swallow the tablet whole, and do not split, crush, chew or dissolve it in any solution [1]. A drop is a dissolved product by design. How the approved tablet’s dose compares with the injection is set out in the systematic review of oral and injected doses.
What the one sublingual study measured
The study randomized Sprague-Dawley rats into five groups [4]. One group had an injection of 0.011 mg/kg. Two had a sublingual suspension at 1 mg/kg, prepared from commercial tablets or from peptide powder. Two had swallowed tablets, at 1 mg/kg and 20 mg/kg.
At 1 mg/kg, total exposure under the tongue was 82.53 ng·h/mL against 15.08 for the swallowed tablet [4]. Relative to the injection, sublingual bioavailability was 0.34% from tablets and 0.29% from powder. The swallowed tablet reached 0.06% at 1 mg/kg and 0.16% at 20 mg/kg.
Three conditions limit what that shows. The subjects were rats, and each received a single dose. The vehicle was a proprietary suspension rather than a retail drop. Every author lists Professional Compounding Centers of America, a supplier to compounding pharmacies [4]. The authors called it a proof of concept that supports research in humans.
What the animal weight data show
One study tested an electrospun sublingual film in a group of 40 female Göttingen minipigs [5]. The pigs were neither diabetic nor obese. Sublingual dosing ran three times a week for 8 weeks, with SNAC or sodium dodecyl sulfate as the enhancer.
Against placebo, the film attenuated weight gain by 24.9% over 9 weeks [5]. The authors report that this exceeded the result in the oral tablet group. They also name the limits themselves: a small group and a short duration. The first author lists a company affiliation, InStar Technologies.
Slower weight gain in growing animals is not weight loss in adults with obesity. The result supports further testing, but it does not describe what a compounded drop does in a person.
What sellers are offering instead
Compounded sublingual semaglutide is common. A review of compounding websites in early 2025 found 17 single-ingredient semaglutide or tirzepatide products [6]. Of those, 82% were sublingual and 18% were orally disintegrating tablets.
The reviewers found a rationale for making such formats. They also wrote that any advantage over the approved oral tablet has not been determined [6]. That matches the record above: the route is plausible, and the human measurement is missing. Whether the injected compounded product produces weight loss is a separate question, answered in whether compounded semaglutide works.
What a reader can check before paying
A drop sold as semaglutide should state three things before it is bought. They are the milligram dose per drop or per milliliter, the compounding pharmacy that made it, and any absorption enhancer in the formula.
If a seller cites a human absorption figure for its drop, the figure should trace to a published study. A PubMed search found none for a sublingual product in people. The safety side of compounded products, including dosing errors, is covered in whether compounded semaglutide is safe. A related category, the patch, is covered in whether GLP-1 patches work.
For background, the adjusted comparison of the approved tablet with the injection is in the bridged-trial comparison of the 25 mg tablet.