Every change, on the record.

We said we found no lookalike shop using the Peptide Partners name. That held for our 9 October 2026 searches, but peptidepartners.lat had been registered on 14 July 2026. The line is now dated and names the domain.

On 29 October 2026 RDAP for peptidepartners.lat gave a registration date of 14 July 2026, and the domain served a Spanish-language page under the shop's name. We describe it as a lookalike domain; we found no record linking it to the shop. No score changed.

The one-line summary for thymosin alpha-1 said it was on FDA's Category 2 list. FDA now lists it as nominated but withdrawn, not Category 2.

FDA's page on bulk drug substances that may present significant safety risks (content current as of 22 April 2026), read on 29 October 2026. The detailed compounding note already said this; the summary line now matches it.

We said Lilly's complaint against American Peptides was not public and that the filing did not confirm which website it concerns. The complaint is now public and names americanpeptides.us.

W.D. La. 5:26-cv-04146, Doc. 1, paragraph 11, read on CourtListener on 26 October 2026. The quick answer, the relief FAQ, the defendant section and an update note now use the complaint. The same confirmation and the Cenexa complaint's allegation (S.D. Tex. 4:26-cv-08806, Doc. 1, paragraph 45) were added to the 11-sellers article.

We described Peptide Partners' certificate ledger as public. From 26 October 2026 the ledger, catalogue and prices require a free account that asks for a research position and institution.

The ledger and shop were open without an account on 19 October 2026. The pro, the findings list and the warning-letter FAQ now give the date of the change. Score unchanged.

SURMOUNT-5 citation year corrected from 2024 to 2025 (NEJM 2025;393:26-36), and its semaglutide arm described as the maximum tolerated dose, 1.7 or 2.4 mg, not 2.4 mg only.

The weight-loss figures (20.2% vs 13.7% at 72 weeks, 751 adults) were already correct and are unchanged. Checked against the PubMed abstract on 22 October 2026.

We priced retatrutide at a median of $8.50 per mg from listings captured in May and June 2026. On 21 October 2026 the median across 20 listings was $6.50, so the monthly figures are now lower.

The quick answer, excerpt, FAQ, cost table and annual totals now use our 20-21 October 2026 capture: 20 listings from 5 vendors, $3.50 to $13.05 per mg, median $6.50 (about $52 to $312 per four weeks across the trial dose steps). The May-June median of $8.50 stays in the table as a dated column. Most of the change comes from which vendors sell retatrutide, not from price cuts: two May-June vendors left the index and two lower-priced vendors joined. The under-fill example now uses the October median ($7.37 per effective mg, was $9.64).

We counted the Astra case as dismissed and said seven of Lilly's 11 retatrutide suits were open. Astra is still open, so eight are. We also said no settlement terms were public for Texas Peptides; Lilly has described terms to the press.

The 2 September 2026 notice in Eli Lilly v. Astra LLC (W.D. Tex. 5:26-cv-05147) ended the claims against Astra LLC only. The case continues against "Astra Peptides", an operator whose identity is unknown: Lilly moved for pre-service discovery on 30 September and the court's 1 October advisory names that defendant. From 10 October we counted the case as dismissed; the correct counts are ten open on 10 October and eight on 22 October. Separately, Becker's Hospital Review reported on 13 October that a Lilly spokesperson told The Washington Post two August cases had settled, including Texas Peptides, which agreed to stop selling, advise customers to stop taking the products and donate an undisclosed amount to America's Poison Centers. The page now attributes that to Lilly and notes the 15 October dismissal notice states no terms. The title, summary, FAQ and August-case table are updated.

We said Lilly voluntarily dismissed the Astra case on 2 September 2026. That notice dropped Astra LLC only; the case continues against the unidentified operator of "Astra Peptides". The lawsuit table now also shows the Texas Peptides and Lone Star dismissals (15 and 16 October).

Our FAQ said Lilly voluntarily dismissed the Astra case on 2 September 2026. Only Astra LLC was dropped; the case is still open against "Astra Peptides". The FAQ now also lists the Texas Peptides and Lone Star dismissals.

Our table listed the Astra case as dismissed on 2 September 2026. That notice dropped Astra LLC only; the case is still open against "Astra Peptides".

We listed a 2 September 2026 notice of voluntary dismissal in the Astra case without saying it dropped Astra LLC only. The case continues against "Astra Peptides"; the line now says so.

We said nine of Lilly's 11 retatrutide suits were open. Lilly has since dismissed two more, so seven are open.

Lilly filed notices of voluntary dismissal with prejudice against Texas Peptides (W.D. Tex. 5:26-cv-05146, 15 October 2026) and Lone Star Peptide (S.D. Tex. 4:26-cv-06562, 16 October 2026), before either seller answered. The notices give no reason. The title, summary, FAQ and the August-case table now show four dismissals and seven open cases.

We listed 1,563 Trustpilot reviews with a five-star aggregate as a point in Modern Aminos' favour. Trustpilot has since removed the profile.

On 19 October 2026 Trustpilot showed the modernaminos.com profile as removed because the business "goes against our guidelines", and no score or reviews are visible. The review figure is out of the pros, the cons and the FAQ say the profile was removed, and the 1,563 figure is kept only as what we recorded at audit. No score changed.

We showed Mile High Compounds' Trustpilot profile at 4.8-4.9 from 316-514 reviews and told readers to check the live profile. Trustpilot has since removed it.

On 19 October 2026 Trustpilot showed the milehighcompounds.is profile as removed because the business "goes against our guidelines". The buyer feedback table and the support line now say so, and the 9 October figures stay as dated history. No score changed.

We said Panda's claimed Trustpilot rating (4.8 from 73) disagreed with the profile, which a cached copy showed with one review and a 3.7 score. Read directly, the profile shows 4.7 from 84 reviews going back to 16 March 2026, close to Panda's claim.

On 9 October 2026 Trustpilot would not open for us and we relied on a cached copy. On 15 October 2026 we read the profile directly, including its last page for the oldest review: 4.7 from 84 reviews, 92% five-star and 1% one-star, the oldest dated 16 March 2026, and the company says it invites reviews. The FAQ and the buyer-feedback table are corrected and the figures now sit in a dated Public reviews section. No score changed.

We said Kovera found a different molecule in three Instant Peptides samples and that Glacier Aminos' three identity flags were all on blends. The reports show identity and net content as "Unknown", and only one of Glacier's three (Glow) is a blend. We also listed Glacier Aminos at 77 certificates; our mirror holds 78.

On 15 October 2026 we re-read the live Kovera reports KVR-2026-CA5B72 (Instant Peptides) and KVR-2026-40AF14 and KVR-2026-2C8A1B (Glacier Aminos): identity and net content print as "Unknown" and no different compound is named. Our copies of these and of KVR-2026-9197F8 and KVR-2026-0815CD (Instant Peptides), all dated 11 May 2026, carry an unconfirmed identity with no measured net content. "GLA-3 RT" and "MTP-31" are Kovera listings for retatrutide and SS-31, single molecules, not blends. Glacier Aminos has 78 certificates in our mirror (19 February to 20 June 2026), not 77. The identity section and leaderboard are corrected; no score changed.

The purity FAQ said tests 'typically show 96–99% purity'. Our pooled lab record shows a median of 99.88% (184 results, lowest 98.38%), none of 176 under 90% of label, and one identity failure. The FAQ is corrected and a section with the figures is added.

The 96–99% range predated our pooled record and was not computed from it. Source: the 2,087-result de-duplicated record behind the October 2026 purity study (Kovera, Janoshik, Particle, Swiss Chems, Ascension, Injectify): tirzepatide 187 results from 91 clients, median fill 111.1%, 99 of 176 over 110% of label. The identity failure is Kovera KVR-2026-17CDBE (20 May 2026).

The purity FAQ said tests 'typically show 96–99% purity'. Our pooled lab record shows a median of 99.92% (41 results, lowest 98.43%), none of 38 under 95% of label, and no identity failures. The FAQ is corrected and a section with the figures is added.

Source: the 2,087-result de-duplicated record behind the October 2026 purity study: semaglutide 41 results from 20 clients, median fill 111.5%, 22 of 38 over 110% of label.

The purity FAQ said tests 'typically show 95–99% purity'. Our pooled lab record shows a median of 99.90% across 92 results (lowest 97.60%), with 1 of 90 under 90% of label. The FAQ is corrected, and the Kovera link now gives the current mirror size, 1,828.

Source: the 2,087-result de-duplicated record behind the October 2026 purity study; MOTS-c 92 results from 65 clients, median fill 109.6%, no identity failures. The related-reading link text said 1,816 Kovera COAs (the 26 May 2026 snapshot); our mirror held 1,828 on 2 August 2026. The Kovera-only testing table said 0 vials were "under labelled content"; four of the 85 measured 94.7% to 98.9% of label, so the row now reads "under 90% of labelled content" (still 0). We also removed a claim that MOTS-c was the second most-tested compound after BPC-157; in the Kovera mirror it ranks fifth and BPC-157 sixth.

The purity FAQ said tests 'typically show 96–99% purity'. Our pooled lab record shows a median of 99.84% across 22 results (lowest 97.28%), and 2 of 21 measured vials under 90% of label. The FAQ is corrected.

Source: the 2,087-result de-duplicated record behind the October 2026 purity study; AOD-9604 22 results from 17 clients, median fill 110.2%, 3 of 21 under 95% of label, no identity failures.

The purity FAQ said tests 'typically show 95–99% purity' and that identity confirmation was usually high. Our pooled lab record shows a median of 99.89% (92 results, lowest 98.22%), 1 of 89 under 90% of label, and two Kovera identity failures in May 2026. The FAQ is corrected.

Source: the 2,087-result de-duplicated record behind the October 2026 purity study; tesamorelin 96 results from 62 clients, median fill 109.1%. Identity failures: Kovera KVR-2026-0815CD (11 May 2026) and KVR-2026-C29EF3 (22 May 2026, 0 mg measured).

We said 175 of 323 retatrutide vials with a measured content held more than 110% of label. That count was not de-duplicated. On our de-duplicated record it is 159 of 307; the figure of 17 above 120% is unchanged.

The old count added the Kovera and Janoshik mirrors without removing Janoshik results issued under several test numbers. The de-duplicated record (323 results from Kovera, Janoshik, Particle, Ascension and Injectify) has 307 with a measured content: 159 over 110% of label, 17 over 120%, 2 under 90%. It matches our retatrutide lab-record analysis and the October 2026 purity study.

The page title and summary named only TRIUMPH-2, but the page covers all four retatrutide Phase 3 trials. The title now reads 'Retatrutide Phase 3 results: TRIUMPH-1 to TRIUMPH-4' and the summary gives each trial's 12 mg figure.

No trial figure changed. The new summary uses the figures already in the page: TRIUMPH-4 28.7% at 68 weeks (Lilly topline), TRIUMPH-1 25.0% (NEJM), TRIUMPH-2 18.8% (The Lancet), TRIUMPH-3 22.6% (Lilly topline).

The FAQ said Prime Research Peptides published per-batch COAs on its product pages. Our May 2026 capture found no COA documents on its sites and no named lab, as the page body already said. The FAQ is corrected. The shop closed in October 2026; the score is unchanged.

The FAQ answer predated the May 2026 catalog capture and contradicted the page body. A "[TBD] state" placeholder in the open-questions list was also removed.

The page said a first-hand audit was in our 'M2 audit-cycle queue' and listed test orders we would place under aliases. vialaudit has placed no test order and has none scheduled. We removed that wording and replaced the section with checks a buyer can make. The score is unchanged.

The removed lines read as if an order were planned. The profile is a public-data audit. The "What we'd verify in a full audit" section is now "What a buyer can check".

The page said a first-hand audit 'lands here when we complete a test order' and that one 'would identify' the testing lab. vialaudit has placed no test order. We removed that wording and replaced the section with checks a buyer can make. The score is unchanged.

Also removed: a line saying per-mg prices needed "a first-hand catalog walk". The US catalog is login-gated, which the page already states.

The page said a first-hand audit was in our 'M1 audit-cycle queue', would 'land here when our audit cycle ships', and listed test orders we would place under aliases. vialaudit has placed no test order. We removed that wording and replaced the section with checks a buyer can make. The score is unchanged.

The FAQ line "First-hand audit content lands here when our cycle completes" was removed as well.

The page said a first-hand audit 'lands here when we complete a test order through EU shipping'. vialaudit has placed no test order. We removed that wording. The score is unchanged.

A second line saying the open question was one "we'd answer with a first-hand order" now points to the vendor's Janoshik reports with live verify links instead.

We removed wording describing what 'a first-hand audit would' do, including a test order and contacting the lab and vendor. vialaudit has placed no test order. The section now lists what public records leave open. The score is unchanged.

Same change, smaller, on Loti Labs, Biotech Peptides, Research Peptide Hub, CertaPeptides, Prime Research Peptides, Ion Peptide, Amino Asylum and Peptide Sciences: "would verify in a full audit" sections renamed "What is still open" and promises of an audit cycle removed. No score changed.

Five pharmacy profiles (Hims, Trillium, Henry Meds, Fifty410, Mochi Health) said scores would recalculate 'when a first-hand consult ships' and listed a consult 'placed by the vialaudit desk'. None is planned. We removed that wording; each section now lists what public records leave open. No score changed.

The same wording was removed from the longevity telehealth ranking, and "audit cycle" promises were removed from three data-synthesis articles (three vendors closed, failure taxonomy, Janoshik 7,164 tests) and the shared-COA-URL article.

We said SS-31 (elamipretide) was not FDA-approved for any indication. FDA granted it accelerated approval as Forzinity for Barth syndrome on 19 September 2025. The quick answer, FAQ and notice are corrected, and a section on its registered trials is added.

The page predated the approval and was not updated when it happened. Our legality tracker already recorded the approval correctly. Source: FDA press announcement, 19 September 2025. The approval is limited to Barth syndrome in patients weighing at least 30 kg, with a confirmatory trial required. Vendor-sold SS-31 is not the approved drug.

We removed Swiss Chems from the ranked vendor list and ended our affiliate link and VIALAUDIT10 code. The profile stays up as a record; the score is unchanged at 46.

On 12 and 13 October 2026 the Swiss Chems referral link (swisschems.is/ref/6723/) redirected to the shop's refund and return policy page rather than the store. The store also carries FDA warning letter 695663 (10 December 2024). We dropped the affiliate link and coupon, marked the vendor as not listed, and updated the disclosure and FAQ on its page. No score changed.

We lowered Ion Peptide's COA-quality subscore from 9.5 to 8.5 (composite 85 to 81). Its own site now returns a site-wide Cloudflare block, so the COA case rests on the lab-held Kovera records rather than Ion's public page.

The 9.5 rested on Ion's public /lab-results page and a two-month Kovera window. On 12 October 2026 ionpeptide.com returned a site-wide Cloudflare 403 to us (every path, from three vantage points), and retail prices are account-gated. The 106 Kovera records are independent of Ion's site and still stand, so we reframed the COA claim to lab-held records and lowered coaQuality from 9.5 to 8.5.

We said Limitless (Limitless Biotech) ships US-only with no international option. Its own shipping policy says it ships worldwide, with customs risk on the buyer, so we corrected the shipping section.

The profile listed shipsTo as US only and said we found no international option. On 12 October 2026 the shop's FAQ and shipping/returns pages state it ships to "virtually any address in the world", with the buyer bearing customs risk, and give international wire instructions in EUR, GBP, CAD and AUD. We corrected the shipping section; international delivery is untested.

We scored Pure Rawz 80/100 and said it had no public regulatory action. It has an open FDA warning letter (715218, 8 Sep 2025) over tianeptine products and has stopped selling, so we marked it retail-ceased and removed the affiliate link.

Our profile rated Pure Rawz 8.0/10 and listed "no public regulatory action against the peptide catalog". FDA's warning-letter database holds letter MARCS-CMS 715218 (8 September 2025, CDER) to Pure Rawz naming five tianeptine products. Separately, by 12 October 2026 purerawz.co had ceased retail and become an affiliate/information site. We set the vendor to shutdown (retail ceased), removed its affiliate link, and corrected the regulatory statement.

We said BioTech Peptides did not surface COAs on product pages. In fact 84 of 89 product pages publish a downloadable per-batch COA (mostly MZ Biolabs), so we raised the COA-quality subscore and the composite to 79.

The profile's central reservation was that testing was "claimed in marketing copy but not surfaced on product pages". A full catalog walk on 12 October 2026 found 84 of 89 product pages carry a downloadable COA, most of them MZ Biolabs reports naming the client and lot, with a set of older SKUs on a lab-unnamed template. We raised coaQuality from 6.8 to 7.5 (composite 7.6 to 7.9).

We had flagged Ascension's code-name GLP-1 SKUs as unconfirmed. We verified them on the live storefront (R-10, T-10, S-5), applied the 0.5 regulatory-exposure deduction, and lowered the composite from 84 to 79.

The code-name-catalogs analysis listed Ascension's coded GLP-1 SKUs as unconfirmed. On 12 October 2026 the live WooCommerce catalog lists R-10/R-30 (retatrutide), T-10/T-30 (tirzepatide) and S-5 (semaglutide), with the real names appearing only inside the COA files. We applied the 0.5 code-name regulatory-exposure deduction, lowering the composite from 8.4 to 7.9.

We listed Kovera's endotoxin reporting as not checked. Its public certificates report an LAL endotoxin result in EU/mL, so we corrected the table and added a section on what those results show.

The lab-price table said Kovera's endotoxin unit was not checked. Our mirror of Kovera's public certificates (2 August 2026) holds 976 peptide certificates with an LAL endotoxin result in EU/mL; 971 read exactly "< 0.10 EU/mL", which we read as a reporting floor. 178 sterility results report no growth after 0 days of incubation. The article now explains both.

We said research-grade GHK-Cu typically tests at 95-99% purity. In 100 public certificates we mirror, the median is 99.88% and none is below 98%. We also fixed a citation that linked to the wrong paper.

The old purity range was not computed from our mirrored records. The page now reports purity and measured content from 102 deduplicated single-compound certificates (Kovera, Janoshik, vendor-published). Our link for Pickart et al. 2015 pointed to an unrelated article (PMC4488188); it now points to PMC4508379.

The Prime Research Peptides page described a live shop. On 12 October 2026 its /shop/ page returned a 404 and the domain listed no products, so we added an offline notice and removed our affiliate link.

The page linked to primeresearchpeptides.com/shop/ with a tracking code and described the root domain as a redirect to Truform Compounds. On 12 October 2026 /shop/ returned a 404, the homepage described itself as an evidence site with no products, and truformcompounds.com showed only a login page for existing customers. The visit link now goes to the plain homepage with no tracking code. The May 2026 score and analysis are kept as the record of the shop at that time.

We said one of Lilly's 11 retatrutide suits had been dismissed and ten were open. Lilly also dismissed the Astra Peptides case on 2 September 2026, so nine are open.

The page listed only Lilly's 31 August amended complaint against Astra LLC (W.D. Tex. 5:26-cv-05147). The docket shows Lilly filed a notice of voluntary dismissal on 2 September 2026. The title, summary, FAQ and docket table now count two dismissals, Astra and Aesthetic Envy. The same status was added on four related pages. We also checked a PacerMonitor listing of the Texas Peptides complaint under S.D. Tex. 4:26-cv-06564; the served case is W.D. Tex. 5:26-cv-05146, the number we cite.

Our methodology, about and home pages described anonymous test orders and blind lab testing as current practice. We have placed no first-hand orders; every vendor score is a public-data audit.

The methodology page, the about page, the home page and our structured data said we place test orders under aliases, ship them to a private mailbox, send vials to two rotating labs that do not know the vendor, log shipping across three orders and test customer service each cycle. We have done none of this. No first-hand test order has been placed with any vendor. Scores come from public lab records, vendor COAs checked against the issuing lab, FDA and court filings, dated shop prices and the user-reported record. The wording on those pages now describes that method and says first-hand orders are a planned step that has not started.

The Limitless Biotech page, and our 12 July 2026 correction, described first-hand test orders and lab results that never happened. They are removed and the score recomputed from public data: 76 to 47.

The page described three April 2026 test orders under aliases, tamper-sealed vials, HPLC results from two labs (96.4% and 96.1% retatrutide, 97.2% tirzepatide, cross-lab agreement within 0.4 points), a shipping log and a customer-service test. Our 12 July 2026 correction repeated that the test order "held up". We placed no order with this vendor and no lab tested its vials for us; those results should not have been published. The page is now a public-data audit built on the shop's published certificate list, Finnrick's 22 tests (7 pass), an Ohio Board of Pharmacy case naming the shop as a clinic's unlicensed DSIP source, our June 2026 price capture and the rebrand record. Subscores moved: COA quality 9.0 to 4.5, pricing 8.0 to 6.0, shipping 9.0 to 5.5, support 6.0 to 5.5, with 0.5 off for regulatory exposure. Composite 76 to 47.

The Particle Peptides page described an April 2026 test order, a four-day delivery and a 27-hour support reply. We placed no order and sent no support question. The score is recomputed from public data: 84 to 62.

The page said we ordered BPC-157, TB-500 and GHK-Cu under an alias on 4 April 2026, received them on 8 April and got a customer-service reply after 27 hours. None of that happened. The page also called the shop crypto-only, said it historically used Janoshik and said 72 batches were in our mirror; its payment page lists card, SEPA and wire, its certificate archive holds Apigenex and Liquilabs reports only, and our mirror holds 79. The page is now a public-data audit of that archive, the shop's published shipping, payment and contact terms, and our June 2026 price capture. Subscores moved: COA quality 8.5 to 6.5, pricing 9.0 to 7.4, shipping 8.5 to 6.5, support 8.0 to 6.5, with 0.5 off for selling retatrutide as "GLP-3". Composite 84 to 62.

Our 22 April 2026 correction for Amino Asylum said we re-tested an independent vial and had recorded a batch at 91.2%. We never tested an Amino Asylum vial.

The 22 April 2026 entry below says a reader's COA contradicted "a batch we recorded at 91.2%", that we "re-tested an independent vial" and confirmed 96.4%, and that we updated a shipping median. We have never ordered from Amino Asylum, which closed in June 2025, and no lab has tested a vial for us. The entry stays in place because this log is append-only; treat its testing and shipping claims as withdrawn. The Amino Asylum page itself makes no testing claim and is unchanged.

The Limitless rebrand article said the shop's certificates were typically Janoshik reports whose batch IDs resolve on the lab's listing. They are not Janoshik reports, and we found none in our mirrors.

The FAQ, the COA section and the alt text of figure 2 described a "Janoshik COA practice". None of the shop's certificates is in our Janoshik or Kovera mirrors, and Peptide Decode reports that the ones it opened came from MZ Biolabs and TrustPointe Analytics. The text and alt text now say so. The diagram image itself still carries the line "Same Janoshik COA practice"; its caption now flags that line as wrong.

Four articles and three peptide guides described independent retesting or cross-lab checks as something our audits do. We have retested no vendor's vial.

The Kovera 1,816-COA analysis referred to "our first-hand test orders". The Janoshik 7,164-test analysis said every audit applies a cross-lab tolerance against independent retesting. "Is Janoshik testing legit?" and "How to verify a Janoshik COA" said our audit cycle retests separate batches. The retatrutide, SS-31 and tirzepatide guides described cross-lab agreement as a factor we weigh or have observed. None of this has happened. Each passage now says we have not retested any vial, and the peptide guides describe what we do weigh: whether a certificate can be checked against the issuing lab's own record.

Every vendor page headed its peptide list "What we tested" with a count of SKUs "ordered", and site labels called shipping days a "median". We ordered nothing; the list is what the shop sells and the days are stated or user-reported.

The vendor template labelled each shop's peptide list "What we tested" and "N SKUs ordered", called its review date "last tested", and showed shipping days as a "ship median". The home page, leaderboard, vendor cards, comparisons and the generated vendor FAQ used the same wording. The list is now "What it sells", the date is "last reviewed", and shipping days are labelled as stated delivery times, with the FAQ noting that we have not timed a delivery. The customer-service subscore description no longer mentions first-response times we did not measure.

Swiss Chems rescored from 78 to 46. Our earlier profile called it a vendor outside US jurisdiction and did not mention FDA warning letter 695663.

The May 2026 preview scored Swiss Chems 78/100 and described a multi-year operating history "outside US jurisdiction". FDA's December 2024 warning letter 695663 addresses the business in New York and names its semaglutide and retatrutide listings; our own 23 June 2026 capture still showed retatrutide on sale. Finnrick passes 4 of 10 samples. The rewritten audit takes 1.5 off for the open letter and scores it 46/100.

We said Swiss Chems removed semaglutide and retatrutide after its December 2024 FDA letter. Our June 2026 capture shows retatrutide and tirzepatide still listed.

Our 23 June 2026 price capture of swisschems.is recorded retatrutide in 5 to 30 mg vials, 18 months after the letter. The semaglutide, retatrutide and tirzepatide pages were gone by 10 October 2026. The sentence now says so.

The comparison misdated sermorelin's withdrawal, called tesamorelin a 29-amino-acid analog, said it could be compounded, and gave unsourced prices.

Earlier versions said sermorelin's approval was withdrawn in 2008 and blamed manufacturing problems. EMD Serono discontinued Geref in 2008; FDA withdrew the approvals effective 18 June 2009 (74 FR 23407) and later found the exit was not for safety or effectiveness reasons. Tesamorelin is the full 44-amino-acid GHRH with a trans-3-hexenoyl group, not the 1-29 fragment. It has been regulated as a biologic since 23 March 2020, so 503A pharmacies cannot compound it; the page said they could. The $5,000+, $300-600 and $80-200 monthly costs had no source and were replaced with a dated Drugs.com cash price for Egrifta WR.

The sermorelin page cited three papers about unrelated topics and gave the wrong Geref approval and withdrawal dates.

The three PubMed links in the earlier citation list led to papers on childhood obesity, vitamin D and deep brain stimulation. They are replaced with the Geref trials and FDA records. The page also said Geref was approved in 1990 for pediatric GH deficiency and withdrawn in 2008. FDA approved it on 28 December 1990 as a diagnostic test and on 26 September 1997 for children with GH deficiency, and withdrew the approvals effective 18 June 2009.

We said Geref was voluntarily withdrawn in 2008. It was discontinued in 2008; FDA withdrew the approvals effective 18 June 2009.

Corrected to cite 74 FR 23407 (19 May 2009), which withdrew NDA 19-863 and NDA 20-443 effective 18 June 2009.

We described TB-500 as a 4-amino-acid sequence. The LKKTETQ fragment we named has seven residues.

The page has been rewritten. Its lab section previously drew on a 200-test Janoshik snapshot with 10 blend tests; it now uses 337 readable Janoshik reports and Kovera certificates, 81 blend results in all. That was a data refresh, not an error.

We said no PCAC recommendation cleared by more than two votes. Our own table showed Semax at 8-5 and Epitalon at 7-4, three-vote margins.

Published summaries of the July 2026 vote disagree on some tallies and FDA has posted no official count. The page now says every published margin was three votes or fewer.

The methodology page listed a 25% label-accuracy weight. Vendor composites are computed from COA quality 40%, pricing 25%, shipping 20% and support 15%; the page now says so.

Since March 2026 the methodology page, the vendor hub and the score grid on each vendor page described the second weight as "label accuracy 25%". No vendor rating has ever carried a label-accuracy subscore: every composite on the site is built from COA quality, pricing, shipping and support, and label accuracy is assessed inside COA quality. We corrected the text and the grid to show pricing at 25%. No vendor score changed.

Panda Peptides rescored from 84 to 77 after we re-read its own lab reports; a 5-Amino-1MQ short-fill claim was withdrawn.

Our earlier profile scored Panda Peptides 84/100. Re-reading eight of the Janoshik reports Panda links on 9 October 2026, we found an AOD-9604 5 mg vial measured at 4.02 mg (80.4%) on Janoshik 212704, which the COA hub's purity headline does not show, and retatrutide sold under the code name "Triple Receptor Agonist". COA quality moved from 9.0 to 8.0 and we took 0.5 off for regulatory exposure, giving 77/100.

We said no tested KPV vial was under its labelled content. Four of 32 measured vials were.

The previous version reported 0 of 33 KPV vials under labelled content and described the set as "none underfilled". A recount of the Kovera and Janoshik certificates (35 certificates, 33 distinct analyses, 32 with a measured content) finds 4 under label. Median purity, 99.93%, is unchanged. The page and its meta description now give 4 of 32.

Kovera pages said 146 vendor-clients and 7 identity-mismatch records. The recount is 145 named clients and 10 mismatch records in the May 2026 snapshot.

Our Kovera analysis, the is-Kovera-legit guide and the vendor leaderboard counted 146 vendor-clients; that total included the bucket of records with no client name. Counting distinct client names printed on the certificates gives 145 named clients, plus 91 records with no name. The same pages said 7 identity-mismatch records; the May snapshot holds 10, and the August mirror 11. The Kovera analysis, the is-Kovera-legit guide, the vendor leaderboard, the COA-revisions piece, the vendor failure taxonomy, the testing-labs comparison, the Janoshik guide and the KLOW read now use 145 named clients.

We placed Lilly's Aesthetic Envy case in the Northern District of California. It was filed in the Eastern District (2:26-cv-02764) and dismissed with prejudice on 11 September 2026.

The August letters article, the where-to-buy-retatrutide guide and the FDA enforcement timeline listed Eli Lilly v. Aesthetic Envy as an N.D. Cal. case. The docket is E.D. Cal. 2:26-cv-02764, and Lilly filed a notice of voluntary dismissal with prejudice on 11 September 2026. All three pages now give the correct court and the dismissal, and link our report on Lilly's 11 lawsuits.

We wrote that Vanguard Laboratory appeared to be an in-house lab name or a private lab with no public verification surface. It is an A2LA-accredited lab with a public verification portal.

Our May 2026 Summit Biotech profile said Vanguard Laboratory "appears to be either an in-house lab name or a private contracted lab without a public verification surface". That was wrong about the lab. Vanguard Laboratory is Olympic Analytical, LLC of Olympia, Washington, holds ISO/IEC 17025 accreditation (A2LA certificate 6377.01, valid to 30 September 2027), and runs a public verification portal at verifiedbyvanguard.com, registered on 27 January 2026, that listed 285 client brands on 9 October 2026. The accreditation scope covers food microbiology, preservatives, trace metals and residual solvents, not peptide purity, assay or identity. Summit Biotech is not a public brand on that portal, so its eight Vanguard reports remain images on its own site that we cannot verify. The profile now says so and links our Hello Stacks audit, which documents the portal.

Our May version said bacteriostatic water was on no federal shortage list. ASHP lists it as a current shortage, and FDA's database showed the main US vial at limited availability in January 2026.

The May 2026 version of this article stated that bacteriostatic water did not appear on any federal shortage list. That was wrong. ASHP lists bacteriostatic water for injection as a current shortage in an entry created 15 July 2025, naming the Pfizer (Hospira) 30 mL vial, citing manufacturing delays and Pfizer as sole supplier. FDA's drug shortage database listed the same Hospira 30 mL vial as "Limited Availability" in an update dated 23 January 2026. The article now states both listings with dates and links, and adds shelf-life guidance and FDA's August 2026 position on bac water sold alongside research peptides.

TRIUMPH-3 and TRIUMPH-4 trial populations were transposed, the TRIUMPH-4 headline result and duration were wrong, and the FDA submission timeline was out of date.

Our article reported TRIUMPH-4 as showing 24% weight loss at 48 weeks in an obesity-plus-comorbidities population enriched for cardiovascular risk. TRIUMPH-4 ran 68 weeks in adults with obesity or overweight and knee osteoarthritis, and showed 28.7% average weight loss at 12 mg. The cardiovascular-disease population is TRIUMPH-3, which we had described as the osteoarthritis trial. The two were transposed. We also listed TRIUMPH-1 through TRIUMPH-4 as published as of May 2026, when TRIUMPH-2 and TRIUMPH-3 did not report until 23 July 2026, and we described a 2026 FDA submission target that Eli Lilly has since specified as a Q1 2027 Biologics License Application. A five-trial programme roster including a TRIUMPH-5 cardiovascular outcomes study could not be verified against Lilly's own announcements and was removed rather than rewritten. The article has been rewritten to cover all four reported readouts from Lilly's topline releases, and the retatrutide peptide page, dosing protocols, where-to-buy and retatrutide-versus-tirzepatide pages were updated to match.

Paradigm Peptides shutdown date corrected from December 2025 to March 2024, the charge description corrected, and unsourced investigation and customer-migration claims removed.

Our profile recorded Paradigm as having ceased operations in December 2025 following a federal plea. Operations actually ceased in March 2024 during the federal investigation, roughly twenty months earlier, with guilty pleas in 2025 and sentencing on 30 July 2026. We also described the charge as concerning SARMs products specifically and the peptide catalog as not being the formal target. Both defendants pleaded to introducing unapproved new drugs into interstate commerce with intent to defraud and mislead, which covers the business as a whole, and the owner admitted forging laboratory certificates across the catalog. Separately, the earlier version carried customer-migration figures and an investigation timeline we could not trace to a primary document. Those passages were removed rather than rewritten. Primary source is the US Attorney's Office for the Northern District of Indiana announcement of 30 July 2026.

Limitless Biotech (formerly Limitless Life Nootropics) composite lowered from 91 to 76 to weight operating-history trust signals, and the profile retargeted to the current brand name.

Our April 2026 first-hand test order held up (96%+ retatrutide purity, clean cross-lab agreement), so COA quality and shipping subscores stayed high. But the composite had not accounted for the vendor's operating history: scam-labelled Reddit threads over billing/support disputes on the legacy brand, a pseudonymous owner-of-record, and a promo-seeded Trustpilot base. Support subscore moved 9.4 to 6.0 and composite 91 to 76 (green to navy). The profile name was updated from "Limitless Life Nootropics" to "Limitless Biotech" to match the current brand.

Refreshed with current data and reviewed for accuracy.

Refreshed with current data and reviewed for accuracy.

Refreshed with current data and reviewed for accuracy.

Refreshed with current data and reviewed for accuracy.

Refreshed with current data and reviewed for accuracy.

Refreshed with current data and reviewed for accuracy.

Refreshed with current data and reviewed for accuracy.

Updated last-tested date and shipping median after a reader-submitted COA contradicted our March finding.

A reader sent in a March 2026 third-party COA showing 96.4% purity on a batch we recorded at 91.2%. Re-tested an independent vial and confirmed the higher figure. Subscore for COA quality moved from 8.1 to 8.5; composite from 83 to 84.

Composite weighting clarified: shipping subscore moved from 25% to 20% to make room for an explicit 25% label-accuracy weight.

Earlier copy of the methodology page implied label accuracy was bundled into purity. We split the two and rebalanced. No vendor scores were retroactively adjusted; the next audit cycle uses the revised weights.