The one sentence summary

There is no FDA-approved generic semaglutide or tirzepatide as of April 2026. The branded patents do not expire in the US until 2031 (semaglutide) and 2036+ (tirzepatide). What exists in the meantime is compounded semaglutide and tirzepatide, produced under specific FDA shortage-list exemptions by 503A (retail compounding) and 503B (outsourcing facilities) pharmacies. That is a legally and clinically different category from a generic. It is cheaper; it is also less regulated, less consistent, and may disappear when the shortage designation is lifted.

The price reality across every access path

Wegovy cash list (Novo Nordisk): $1,349/mo. Almost nobody pays this. Inflation-adjusted GoodRx rarely beats $1,100.
Zepbound cash list (Eli Lilly): $1,086/mo. Same story.
Wegovy / Zepbound insured with manufacturer savings card: $0–$50/mo copay if commercially covered. Federal program patients (Medicare, Medicaid, VA) ineligible for savings cards.
LillyDirect Zepbound single-dose vials: $349–$499/mo depending on dose (2.5 and 5 mg) as of 2026. Requires online consultation; cash pay only.
NovoCare Wegovy vials: $499/mo for all strengths as of 2026. Cash pay only; 2.5-month supply commitment typical.
Compounded semaglutide (Hims, Henry Meds, Mochi, local compounding pharmacies): typically $199–$349/mo for semaglutide; $299–$499/mo for tirzepatide. Prices dropped in 2025 as more 503B facilities entered.

What “compounded” actually means, legally

Pharmacy compounding is the practice of combining, mixing, or altering drug ingredients to create a medication tailored to an individual patient. Federal law carves out two categories:

503A pharmacies: state-licensed retail compounding pharmacies. Produce one patient at a time against a specific prescription. Quality oversight varies by state board of pharmacy. Most telehealth compound semaglutide comes through 503A channels.
503B outsourcing facilities: federally registered, FDA-inspected, must meet cGMP (current Good Manufacturing Practices). Can produce in larger batches. Used by hospital systems and some premium telehealth brands.

A 503B product is produced under stricter quality oversight than a 503A product but neither is an “FDA-approved drug” in the sense that Wegovy or Zepbound is. The FDA does not review efficacy data for compounded products; it does not approve labeling; it does not require bioequivalence testing.

The FDA shortage-list mechanism — and why the compounded market exists

Under section 503A of the Federal Food, Drug, and Cosmetic Act, compounding a “copy” of an FDA-approved drug is prohibited — unless that drug is on the FDA Drug Shortage List. Both semaglutide (October 2022) and tirzepatide (October 2023) were added to the shortage list, which legally opened the door for large-scale compounding.

In October 2024, the FDA declared the tirzepatide shortage resolved. Eli Lilly and the FDA faced a coordinated legal challenge from Outsourcing Facilities Association, and a court briefly paused enforcement. By early 2025, the tirzepatide compound window had effectively closed. Semaglutide shortage was declared resolved in February 2025 with a similar 60–90 day wind-down. As of April 2026, most compounded GLP-1 is either (a) sold as “personalized formulations” with added ingredients like vitamin B12 or cyanocobalamin to argue medical necessity and clinical individualization, or (b) sold in legal gray zones that carry genuine regulatory risk to both pharmacy and prescriber.

The efficacy question: does compounded work as well?

No head-to-head RCT has been published comparing branded and compounded GLP-1. What exists is:

Manufacturer testing data: Novo Nordisk and Eli Lilly have tested samples of compounded product purchased from telehealth providers. Reported findings included under-potency (50–80% of label strength), over-potency, sterility concerns, and presence of excipients not in the branded product.
Patient-reported outcomes in communities: Reddit and Facebook cohort data suggest real-world TBWL on compounded semaglutide averages 60–80% of branded at the same nominal dose. A patient on 2.4 mg compounded may effectively be getting 1.2–1.8 mg equivalent.
FDA adverse event reporting: hundreds of compounded-GLP-1 adverse events reported through FAERS, including dosing errors (vials labeled in mg rather than units), bacterial contamination, injection site reactions, and severe GI symptoms. Most resolved; some hospitalizations.

The honest summary: average efficacy is probably 20–40% below branded, variance is much higher, and the safety downside is non-trivial but small in absolute terms. For patients with no access to branded (no insurance, no savings card, cannot afford LillyDirect/NovoCare), compounded is a defensible bridge. For patients who can afford LillyDirect or NovoCare vials at $499, the economics have tilted back toward branded.

What the trials never tested

The entire efficacy base for GLP-1 anti-obesity therapy — STEP 1 (N=1,961), STEP 2 (T2D, N=1,210), STEP 3 (semaglutide + ILI, N=611), STEP 4 (withdrawal, N=803), STEP 5 (2-year durability, N=304), SURMOUNT-1 (tirzepatide, N=2,539), SURMOUNT-2 (T2D, N=938), SURMOUNT-3 (tirzepatide + ILI, N=579), SURMOUNT-4 (withdrawal, N=670), SURMOUNT-5 (head-to-head, N=751), SELECT (N=17,604), FLOW (N=3,533), STEP-HFpEF, SURPASS-CVOT (ongoing) — every single one of these trials used branded, manufacturer-produced, FDA-registered product. No compounded semaglutide or tirzepatide has been studied in any phase 2 or phase 3 randomized trial. When a telehealth platform markets compounded semaglutide by citing “STEP 1’s 14.9% TBWL,” it is extrapolating — reasonable extrapolation at labeled dose with quality 503B product, but extrapolation nonetheless. Variability in compounded potency, injection technique, and concurrent excipients means real-world mean TBWL on compounded is materially below STEP 1.

Where compounded still makes sense in 2026

Bridge during a prior auth denial: $299 for 2–3 months while you appeal a commercial denial is reasonable. See the prior auth likelihood tool.
Dose the manufacturer does not sell at discount: Zepbound 7.5 mg, 12.5 mg vials are not always available through LillyDirect; some patients use compounded for intermediate doses.
Personalized formulations: B12-added or micro-dose semaglutide (0.125–0.25 mg) for maintenance in patients who cannot tolerate standard labeled doses. See the maintenance dose tool.
Patients without any insurance path: $299 cash for compounded may be the only realistic option versus stopping therapy entirely.

Where compounded does not make sense

If you have commercial insurance that covers Wegovy or Zepbound at < $100/mo: the cost arbitrage disappears.
If you have history of pancreatitis, gallstones, MEN2, medullary thyroid cancer family history: the labeled product has risk mitigation; compounded varies.
If SELECT CV indication is the reason for therapy: the FDA-approved cardiovascular risk-reduction labeling is specific to Wegovy. Compounded is off-label for CV indication.
Pregnancy, lactation, or planning pregnancy within 2 months: contraindicated across the board.

SELECT and the “indicated product” problem

SELECT’s 20% MACE reduction in adults with established CVD was specifically on semaglutide 2.4 mg (Wegovy). FDA added the cardiovascular risk-reduction indication to Wegovy’s label in March 2024. Compounded semaglutide, however chemically similar, is not Wegovy and is not labeled for CV risk reduction. The implications: (a) Medicare Part D coverage expansion for cardiovascular risk reduction applies only to Wegovy, (b) a clinician cannot technically prescribe compounded semaglutide “for cardiovascular risk reduction under the SELECT indication” because that indication attaches to a specific FDA-approved product, (c) patients who need the SELECT-qualifying documented indication for insurance should be on Wegovy proper. For patients whose motivation is weight loss and whose CV risk is not yet in the SELECT cohort, compounded can be a reasonable bridge; for patients who need the cardiovascular labeled indication, it is not a substitute.

How to evaluate a compounded pharmacy (if you go this route)

Ask whether it is 503A or 503B. 503B is materially higher quality oversight.
Ask for a certificate of analysis (COA) on each lot — confirms potency, sterility, endotoxins.
Prefer semaglutide sodium over semaglutide base. The sodium salt was historically preferred for stability; the FDA has flagged semaglutide base as non-approved chemical form.
Verify the state board of pharmacy license. Both the pharmacy and the prescriber must be licensed in the state you receive the prescription.
Avoid “peptide research” suppliers selling “not for human use” semaglutide online. Different regulatory regime entirely; sterility and identity are unverifiable.

FAQ

When will a true FDA-approved generic semaglutide arrive?

Semaglutide US patent coverage extends into 2031–2033 depending on formulation. Canada’s patent expires earlier (2026). Generic approval requires a branded to go off-patent and then an ANDA bioequivalence submission. Realistic US generic timeline: 2032–2034.

Is tirzepatide still compounded in 2026?

Legally, no — the shortage was resolved in October 2024. In practice some 503A pharmacies continue under “personalized formulation” arguments, which invites FDA enforcement. Risk has increased in 2026.

What is “semaglutide base” versus “semaglutide sodium”?

The FDA-approved branded product is the sodium salt. “Semaglutide base” is a chemical variant that the FDA has flagged as not covered by the compounding rules. Some compounders substituted base for sodium to argue they were making a different drug — FDA disagreed.

Can compounded product cause pancreatitis?

Any GLP-1 agonist can cause pancreatitis; labeled rate is about 0.1–0.2%. Compounded product may add variability from excipients or over-potency. Symptoms: severe upper-abdominal pain radiating to back, vomiting. ER evaluation mandatory.

Do compounded products work for T2D glycemic control?

Probably, at reduced potency. But no CVOT or glycemic-endpoint trial has been done on compounded product. For A1C-driven decisions in T2D, branded is clearly preferred. See A1C reduction projector.

Generic vs brand GLP-1

Your inputs

Results