Clinical Reference & Evidence
H

Hormonaly

Hormonaly sells Helix, an evidence engine for peptide, hormone, endocrine and longevity medicine, to clinicians and to the companies that build clinical software for them. More than 40 specialized agents retrieve studies from six biomedical databases, PubMed among them, rerank them, draft an answer and verify the citations before and after drafting. Each claim carries a GRADE evidence rating from A to D, and the shared library holds more than 590 graded protocols. A nightly job folds in new research, FDA actions and trial updates; routine refreshes apply automatically, and a physician signs off on changes to clinical guidance.

Individual clinicians use Helix and the full Workspace free, in English or Arabic. Organizations license the Helix API from $499 a month for 5 million tokens, with Rx, Scribe and Protocol endpoints and an MCP server, or deploy it white label in their own cloud or on premises. Celia Group runs its AI prescribing platform and EMR at celiaone.ai on Helix and PharmacyOS inside Canvas Medical, where the prescriber signs every order. Science & Humans, a Canadian hormone health platform, embeds the API, and the Hospital for Endocrine Surgery uses its review agents to screen and grade surgical literature.

Endocrinologist and geneticist Fady Hannah-Shmouni founded the company in 2025, and Omar Saleem leads AI and education. Hormonaly counts more than 30 deploying organizations and clinicians in five countries. Its flagship is now Perceptors LMS, an academy platform sold under its own brand, with Helix supplying the evidence beneath every course.

AI Health Index verifiedNovember 19, 2026
Compare Hormonaly with other vendors
Founded
2025
Headquarters
Washington, DC
Website
hormonaly.ai/
Categories
clinical-reference-and-evidence, clinical-decision-support
Indexed Products
Helix API, Hormonaly Workspace, E-prescribing module, Hormonaly-1 LLM, Hormonaly Evidence Runtime
Buyer Segments
Clinicians in peptide, hormone and longevity medicine, Digital health product teams, Health systems, Translational research teams
Assessment

Capability Axes

An AI Health Index grade measures what a buyer can verify from public sources on the date shown. It is not a rating of how good the product is. A vendor can build an excellent system and grade low on an axis because it publishes nothing an outsider can check. How grades read

AI Capability
BB on AI CentralityThe model is the engine of a core module. The platform carries other value, but this capability does not exist without it.
Vendor Published

Helix is a multi agent pipeline that turns a clinical question into a graded, cited answer. A model coordinates retrieval across six biomedical databases, reranking and synthesis, while separate checks tie each statement back to its source and attach a GRADE rating from A to D. The protocol library also works as a reference on its own: more than 590 graded protocols and compound entries a clinician can browse without asking a question.

Workspace adds case notes, compound comparisons, dosage tables and the Three Lens analysis, and the API puts the same engine inside a partner's product. The library carries reference value without a model, but synthesis, protocol drafting from a case and the prescribing agents all depend on the models that assemble each response.

BB on Autonomy and Oversight ModelThe oversight structure is described and one part is missing, commonly the threshold at which the system stops or what happens after it is wrong.
Vendor Published

Oversight differs by surface. In the evidence pipeline, a nightly job scans new research, FDA actions and trial updates; routine updates apply automatically, while a safety signal, a change in evidence certainty or a regulatory change routes to review, and a physician approves any update that changes clinical guidance.

Answers carry the condition that every output requires clinician review before clinical use, and the terms prohibit autonomous prescribing or dispensing and direct to patient guidance outside a licensed care relationship. In the Canvas Medical prescribing workflow the prescriber reviews and signs every order before it is sent, and Supercell's evidence copilot requires physician sign off on each protocol draft. Through the plain API, Helix returns evidence to a partner's application, and how review is enforced there is the partner's design.

BB on Model and Technology TransparencyThe approach or the suppliers are named without the version and update discipline behind them. Naming a supplier is the entry to this band both here and on Model Supply Chain Disclosure, which ask different questions of the same disclosure: who receives the data, and what produces the output.
Vendor Published

The Helix router can call Hormonaly's own clinical model and models from Anthropic and OpenAI, with fallback between providers. For private licensing, the company offers Hormonaly-1, a clinical language model fine tuned on its graded evidence material. The pipeline retrieves from six biomedical databases, reranks sources, drafts, and runs a second citation check after generation, a design the company calls dual layer verification.

The evidence library refreshes nightly without retraining a model, and the system is described as model agnostic. The company commits to posting model and methodology changes on its updates page. Which Claude and GPT versions a given request reaches, and when the routing changes, are not identified, so a buyer cannot tie an answer to a particular model release.

BB on Model Supply Chain DisclosureSubstantial partial disclosure, or a chain that is structurally short: an in house build, a cleared model that cannot be quietly swapped, or a deployment where the transfer does not occur at all. Naming the model provider exits the band below into this one; the axis rises from here on the completeness of the party list and on the terms that govern data once it arrives.
Vendor Published

The privacy policy names Anthropic and OpenAI as AI processors, alongside Google Cloud for hosting, Stripe for billing, Resend for email and Google Workspace for internal operations, and states that queries are processed to answer them without training third party models. The hosted service runs on Cloud Run and Cloud SQL in Google Cloud's central US region, with the database reachable on private IP only.

Enterprise clinical deployments add Canvas Medical as the EHR and infrastructure layer, and on those deployments the model providers are covered by business associate agreements. A customer can also keep inference inside its own cloud or data center. Retention terms at each model provider are left to the deployment agreement rather than set out processor by processor.

CC on Clinical and Operational EvidenceNamed customers, or vendor reported percentages with no method, denominator or reference standard. Scale of use is recorded here and is not treated as evidence of benefit.
Vendor Published

The named deployments show where Helix runs rather than how well it performs. Celia Group launched its AI prescribing platform and EMR at celiaone.ai on Helix and PharmacyOS and runs it in production. Science & Humans embeds the API in its Canadian hormone health platform, the Hospital for Endocrine Surgery runs Helix's review agents on surgical endocrinology literature, and NU Pro joins multi omics readouts to graded protocols.

The company counts more than 30 deploying organizations and clinicians in five countries. The randomized trials and reviews inside the library assess the therapies a protocol covers, not Helix itself. There is no citation error rate, no agreement study against clinicians and no measured change in prescribing or outcomes at any named customer.

AA on AI Safety and PHI StewardshipRetention windows, training use and de identification are all stated specifically enough to be contradicted. Safety engineering, meaning guardrails, hallucination mitigation and how a safety event is handled, does not gate this grade: no vendor in this index publishes it, so every note on this axis states that gap rather than leaving it implied.
Vendor Published

Protected health information is confined to enterprise deployments under a signed business associate agreement. The free Workspace and the Starter and Advanced API plans take no patient identifiers, and API partners must send deidentified queries at the level of compound and indication. Workspace chat context lasts only for the session unless a user saves it, and saved sessions are encrypted with AES 256 and deletable at any time.

Partner data and audit logs are kept 12 months by default, enterprise contracts can set custom retention, and deletion requests go to a published address. The privacy policy rules out training third party models on queries and states that data is never sold. Citation and dose checks run before delivery and reported content errors are triaged, though no incident procedure for an unsafe clinical answer is set out.

Regulatory and Compliance
BB on HIPAA and BAA PostureBusiness associate status is stated and supported by a substantive privacy document, with the agreement or its scope not fully published. For a vendor outside the United States, an equivalent regime documented to this depth grades here. A vendor whose product does not process protected health information grades here too when it says so plainly and explains the scope, with a service privacy document behind it: stating a position a buyer can rely on is the posture this axis grades, and the band above is closed to it because there is no agreement to publish.
Vendor Published

Business associate status is tied to one tier. Enterprise deployments can process protected health information under a signed agreement, on HIPAA, SOC 2 and HITRUST certified infrastructure operated with Canvas Medical, while the Free, Starter and Advanced plans exclude patient identifiers and carry no compliance claim.

The trust center, updated August 4, 2026, sets out that policy plan by plan on one page, and the privacy policy describes the standard product's limits separately from the Google Workspace agreement that covers internal operations. The agreement itself is requested from the team and scoped per engagement, so its subcontractor terms, and how it reaches the model providers named in the privacy policy, are confirmed only at contracting.

DD on Security Certifications and Trust CenterControls are asserted with nothing independent behind them, or nothing is published. Read the note before concluding anything: this is the grade most often corrected on a second pass, because assurance material frequently sits on a parent domain or inside an old announcement rather than on the product pages.
Vendor Published

The hosted service runs TLS 1.3 in transit and AES 256 at rest on Google Cloud, with ephemeral Cloud Run containers, a Cloud SQL database reachable only on private IP, Google OAuth sign in, role based access control, rate limiting on every endpoint, tamper evident audit logs and API keys hashed with SHA 256. Periodic reviews report zero high severity CVEs.

The certifications on offer belong to partners: Google holds SOC 2 Type II and ISO 27001 for its infrastructure, and Canvas Medical's HITRUST, SOC 2 and HIPAA documentation covers enterprise clinical deployments. Hormonaly holds no attestation of its own, and the security package carrying the Canvas documents is released on request rather than through a self service portal.

BB on FDA and Regulatory StatusThe pathway is stated and in progress, or a clearance is named without the vintage and scope a buyer needs to match it to the product on offer. A product outside device regulation also sits here when the vendor publishes what it is for and in what setting and the product on offer matches, or names another regime it works under, such as a registration outside the United States. A general clause saying the service is not medical advice does not reach this band.
Vendor Published

The service terms place Helix outside device regulation as evidence synthesis for licensed professionals, and the trust center describes it as a decision support and evidence synthesis tool whose output informs professional judgment and always requires clinician review.

The prohibited uses are explicit: no substitute for medical advice, diagnosis or treatment, no autonomous prescribing or dispensing, no direct to patient guidance outside a licensed care relationship, and for US partners, sourcing guidance limited to FDA registered 503A and 503B compounding pharmacies. The product stays close to that line.

Workspace can assemble a structured protocol from case details, and the Canvas workflow drafts prescriptions for a prescriber to sign, which keeps the clinician's review at the center of the position. No specific statutory exclusion is named for the current workflows.

BB on AI Governance and Bias DisclosureA governance framework with named process behind it, such as certification to an artificial intelligence management standard, or material written for a customer own review committee to evaluate the product with. A framework covering the vendor whole portfolio counts when its scope reaches this product; what it cannot supply is the product specific evaluation the band above asks for. Published results within the one group where the performance of this product is most at risk also sit here: evaluation on the question this axis asks, short of the comparison across groups the band above requires. A medical device clearance or certification, and the quality system behind it, does not reach this band on its own; it is graded under regulatory status. A regulator reviewed plan governing how the model may change does reach it.
Vendor Published

An independent Medical and Scientific Advisory Board oversees the evidence record. The pipeline applies the Cochrane risk of bias framework to studies and GRADE certainty when it synthesizes a protocol, checks citations against their sources and labels every claim A to D. A physician reviews changes that affect guidance before they enter the shared library, user reported corrections go through the same governed update process, and enterprise customers can keep their own formulary and local deviations inside their tenancy.

That governs how evidence becomes an answer and who authorizes a change. It leaves open how answer accuracy and scope control compare between English and Arabic, and across the patient populations served by partners in Canada and the Gulf.

CC on AI Liability and RecourseMechanisms exist that let someone challenge an output, such as audit trails, source traceability or review before commit, with nothing standing behind the output and no route for the harmed party.
Vendor Published

A clinician can open any citation, compare the answer with its source and report an error in a protocol, monograph or citation by emailing the page link and the issue. The clinical team triages each report, corrections ship through the governed update process, and material changes are noted publicly. A prescription draft in the Canvas workflow stays editable until a licensed prescriber signs it. Those routes correct the reference content.

The service terms disclaim warranties of accuracy, completeness and practical applicability and exclude indirect and consequential damages where the law allows. The 99.95% Enterprise SLA covers availability rather than answers, and no remedy attaches to an individual clinical output once it has been used.

Integration and Deployment
BB on EHR and Interoperability DepthNamed systems with read access or one directional writing, or standards support with named deployments behind it.
Vendor Published

Helix returns evidence and structured responses over a REST partner API, with an MCP server on the higher tiers, and runs inside healthcare applications. The named EHR route is Canvas Medical: enterprise implementations use its SDK, plugin framework and FHIR API, with Canvas authentication, chart access and plugin effects covered in the integration material.

Celia Group's live platform uses Helix and PharmacyOS for prescribing inside Canvas, where a draft order is written against graded evidence, carries compound level dosing, interaction screening and regulatory standing, and is signed by the prescriber. The Prescriber Portal reads compound, interaction and regulatory data straight from the evidence record. Exactly which chart fields Helix reads and writes in Celia Group's deployment is not set out.

BB on Deployment Model and Data ResidencyOptions and residency are stated with isolation or the processing path left open.
Vendor Published

The default deployment runs on Google Cloud in the central US region, on ephemeral Cloud Run containers with no persistent disk and a private Cloud SQL database. Enterprise licenses can run the Helix API in the customer's own VPC or on premises for zero data egress, with the design scoped per engagement. Clinical deployments built with Canvas Medical use dedicated application containers and a separate database for each customer in Canvas's AWS environment in the eastern US.

Residency arrangements for EU and GCC customers are offered by scope, and the hosted service targets 99.9% uptime, with a 99.95% SLA on Enterprise. The inference path is what stays open: where model calls go for each enterprise configuration is set in its agreement.

Commercial
AA on Commercial TransparencyPublished tiers with figures, a stated unit of charge, and a route to start without a sales conversation.
Vendor Published

Pricing is public, with self service at the low end. Individual clinicians get Helix and the full Workspace free, with no card, seats or token meter. Helix API Starter costs $499 a month for 5 million tokens, three API keys, the Rx, Scribe and Protocol APIs and five team seats, with overage at $100 per million tokens and a 10 million token hard cap. Advanced costs $1,999 a month for 25 million tokens, ten keys, MCP access, priority support and 25 seats, with overage at $80 per million.

Enterprise lists $4,999 a month for 75 million tokens with no hard cap, custom overage, white label, on premises or VPC deployment, a 99.95% SLA and unlimited seats. Input and output tokens count together, and PharmacyOS is licensed separately.

CC on Setting and Specialty CoverageCoverage is claimed broadly without specifics, or stated clearly with nothing validating it yet.
Vendor Published

Helix is built for peptide, hormone, anabolic, endocrine and longevity medicine. Its intended users are licensed clinicians, compounding pharmacies, healthcare organizations, educators and researchers, and the partners show the spread: an electronic prescribing EMR (Celia Group), a Canadian hormone health platform (Science & Humans), a surgical endocrinology review engine (Hospital for Endocrine Surgery), a precision health program builder (Supernatural), a peptide multi omics platform (NU Pro) and a performance medicine research tool (Anabol).

The interface runs in English and Arabic, and clinicians use it in five countries. Each setting comes with the job the agents do there. None has a performance result of its own, whether in endocrinology, longevity practice or systematic review.

Alternatives

Alternatives to Hormonaly

Other products that do the same kind of work as Hormonaly, and what each one does.

Real world evidence generation, spun out of Stanford's Green Button project in 2020 and built to answer the question a clinician actually has at the bedside when no trial covers their specific patient.

Clinical search and answer platform for UK clinicians and students, from Medwise AI Ltd, registered in England and Wales and based in Leeds, having started in Cambridge.

Clinical AI reference platform from CaryHealth (Cary Rx Incorporated, Washington, DC), launched in August 2024 on the back of the company's GalenAI acquisition and available at askclair.ai, on the web and as iOS and Android apps, with an API for embedding.

ClinicalKey AI is a generative clinical decision support layer built on ClinicalKey, Elsevier Health's point of care content platform.

Frekil builds AI agents that run observational real world evidence studies for pharmaceutical, biotech and contract research teams on data they already hold or license, from protocol through final report.

FDB (First Databank) supplies the drug knowledge built into the prescribing, pharmacy and claims software that clinicians, pharmacists and health plans use every day.

Commercial

Pricing

Vendor-published figures are labeled as such. Figures labeled “Estimated” are derived from third-party sources and have not been confirmed by the vendor.

Entry Price Pricing Basis BAA Tier Implementation Source
Free
$0 baseline
Workspace is free for individual clinicians. Helix API Starter is $499 per month with 5 million tokens, Advanced is $1,999 per month with 25 million, and Enterprise lists $4,999 per month with 75 million. Overage is $100 per million tokens on Starter and $80 per million on Advanced; Enterprise overage is priced by volume. A business associate agreement is available on Enterprise deployments, on infrastructure operated with Canvas Medical. Free Workspace and the Starter and Advanced API plans exclude protected health information. Starter is self service and Advanced includes guided onboarding. Enterprise integrations, private infrastructure and onboarding are scoped to each deployment, with no fixed setup charge in the plan table. Vendor Published

Starter includes three API keys and five Workspace seats, Advanced ten keys and 25 seats, and Enterprise unlimited keys and seats. Starter stops at a 10 million token hard cap and Advanced at 75 million. The API meters input and output tokens together, annual contracts are available, and PharmacyOS carries its own hosted application license.