PrivacyPolicy
What ZenLLM stores, what it never stores, and how account data is protected.
Last updated
Contents
Who controls your data
ZenLLM, a sole-proprietorship business operated from India, is the controller of the personal data described here. For anything in this policy, including access, correction or deletion requests, ask on Discord at zenllm.org/invite/discord.
What we collect
Account data. Your email, optional display name and username, a password hash, and OAuth provider identifiers when you sign in with Google or Discord. Email, name and IP addresses are encrypted at rest; the email and username are also stored as keyed HMAC hashes so they can be looked up without storing them in the clear.
Usage data. Per-request metadata needed for billing and reliability: model id, token counts, cost, latency, success or failure, and timestamps. Payment records for the transactions you make.
Account-security data. Session dates, an encrypted sign-in IP and a keyed IP hash, browser and operating-system details, a hashed first-party device marker, and bounded display and language settings supplied by your browser. These detect new logins, prevent repeat free-trial signups and support dispute investigation. We do not collect local network addresses through WebRTC.
Support and legal data. Correspondence you send us, and records needed to handle refunds, chargebacks, abuse reports and legal claims.
Prompts and model traffic
ZenLLM routes API prompts to the model service that handles each request and forwards its output to you. We use official provider APIs where applicable, but a request may also pass through another third-party model service. We do not save API prompts or outputs as conversation history on ZenLLM.
We do not log your prompts or completions on our side. We cannot promise zero data retention (ZDR) on every route, because the developers who make these models do not offer it either. A request goes to the model developer’s own API and is handled under that developer’s policy. Read the policy of the model you plan to use, such as OpenAI or Anthropic, and use that model through ZenLLM only if you accept it. If you do not accept the practices of the service handling a request, do not send secrets, personal information, proprietary code or other confidential data through ZenLLM.
Caching follows each model service's policy. We do not sell personal data, run ads, or use third-party analytics or advertising trackers.
Why we use it, and our legal bases
- Contract: to create your account, serve requests, meter usage, process payments and provide support.
- Legitimate interests: to keep the service secure and available, prevent fraud and abuse, debug faults, and enforce our terms. We weigh these against your rights and keep the processing proportionate.
- Legal obligation: to keep records required by tax, accounting, consumer-protection and anti-fraud law, and to respond to lawful requests.
- Consent: for anything optional we ask you about specifically. You can withdraw consent at any time without affecting earlier processing.
Where your data goes
To serve a request, its content is transmitted to the model service that handles it; processing there is governed by that service's policies. When web search is enabled in Chat, search queries are also sent to the search provider and source excerpts are included with the model request. Browser code previews run locally in an isolated environment.
Our database is hosted in the European Union. Some providers we use are located in other countries, including the United States, so your data may be transferred outside your country. Where the law requires it we rely on appropriate safeguards, such as standard contractual clauses, or on the necessity of the transfer to perform our contract with you.
How we protect it
These are the controls actually in place today, described plainly rather than as certifications we do not hold:
- Data is encrypted in transit with TLS, and sensitive account fields are encrypted at rest with AES-256-GCM.
- Emails, usernames and IP addresses are stored as keyed HMAC-SHA512 lookup hashes where we need to search them, so the plaintext is not used for lookups.
- Passwords are stored as salted Argon2id hashes, never in a form we can reverse.
- API keys are stored only as SHA-256 hashes; a leaked database does not yield working keys.
- Sessions use a first-party, host-locked cookie, are short lived, and can be revoked from the dashboard.
- Access to production systems is limited to the operator, and administrative actions on payment and credit records are logged.
- Login and registration are rate limited and abuse-guarded, and repeated failures are recorded against keyed IP and device markers.
No service is perfectly secure. We cannot guarantee that unauthorised access will never happen, but we will notify you and the relevant authority without undue delay if a breach affects your personal data and the law requires it.
We do not hold SOC 2, ISO 27001, PCI DSS or HIPAA certifications, and we do not display badges for certifications we have not earned. Do not treat ZenLLM as suitable for regulated data such as protected health information or cardholder data; those categories must not be sent through the service.
What has been checked independently, and where you can verify it yourself:
- TLS configuration graded A+ by Qualys SSL Labs for both zenllm.org and api.zenllm.org. run the scan.
- HSTS with includeSubDomains and preload, so a browser cannot be downgraded to plain HTTP for any ZenLLM host.
- Strict Content-Security-Policy with frame-ancestors 'none', and a Permissions-Policy that disables camera, microphone, geolocation and payment.
- nosniff, frame denial and a strict referrer policy on every response, and no third-party analytics or advertising scripts.
- A published security.txt at /.well-known/security.txt so researchers can report a vulnerability without guessing how to reach us.
How long we keep it
Account data is kept while your account exists. Expired and revoked session records are removed after 180 days. Usage and payment records are kept for billing history and as required by tax and accounting law. A keyed email or device restriction may remain after a disputed account is deleted, so that a chargeback ban cannot be cleared by re-registering. Deleting a saved conversation removes its content from active storage; backups expire under our retention policy.
Your rights
Depending on where you live, you may have the right to access your personal data, correct it, delete it, restrict or object to processing, receive it in a portable format, and withdraw consent. You also have the right to complain to your local data-protection authority; we would appreciate the chance to resolve it first.
To exercise any of these, message us on Discord at zenllm.org/invite/discord. We will respond within 30 days and may need to verify your identity first. These rights are not absolute: we can refuse a request where the law requires us to keep the data, for example billing records we must retain.
We do not sell or share personal data for cross-context behavioural advertising, so there is nothing to opt out of under "Do Not Sell or Share My Personal Information" laws. We do not knowingly collect data from anyone under 18.
Changes to this policy
We may update this policy as the service changes. Material changes will be announced on the website or by email before they take effect. The date at the top always shows the current version, and earlier versions are available on request.
Contact
Privacy questions, data requests and complaints: join zenllm.org/invite/discord to reach the team. If we cannot resolve your complaint, you may escalate it to the data-protection authority in your country.