Acceptable use.
Build useful, lawful software with Model.sale. Do not use the API, Playground, dashboard or payment flows to harm people, evade safeguards, attack systems or hide abusive activity.
1. Scope and baseline
This policy applies to every request made with a Model.sale account or API key, including requests from the Playground, scripts, agents, integrations and third-party applications you operate. You remain responsible for your users, prompts, outputs, retries and any downstream action your software takes.
Use the service only where you have the right to process the input and where the model output is appropriate for the task. Follow applicable law, the Terms of Service and any rules that apply to the content or industry in which you deploy.
2. Prohibited activity
You may not use Model.sale to:
| Area | Examples |
|---|---|
| Fraud and deception | Phishing, impersonation, scams, fake reviews, financial fraud, credential theft or evasion of identity and payment checks. |
| Cyber abuse | Malware, ransomware, credential stuffing, exploit delivery, unauthorized scanning, persistence, exfiltration, DDoS or attacks on the gateway, payment callbacks or another customer. |
| Harassment and harm | Targeted harassment, stalking, threats, non-consensual intimate content, doxxing or instructions intended to facilitate physical harm. |
| Minors | Child sexual abuse material, grooming, sexual exploitation or any attempt to sexualize or endanger a minor. |
| Privacy violations | Unlawful surveillance, deanonymization, bulk profiling, doxxing or processing sensitive personal data without an appropriate legal basis and authorization. |
| Manipulation and spam | Botnets, bulk unsolicited messaging, fake engagement, election manipulation, coordinated deception or content designed to evade platform safeguards at scale. |
| Service abuse | Resource exhaustion, deliberate stream flooding, abusive retries, account farming, key sharing, payment replay or attempts to bypass RPM, TPM, concurrency or spend controls. |
3. High-impact and sensitive decisions
Do not rely on model output as the sole basis for decisions about employment, credit, housing, insurance, education admission, healthcare, legal status, access to essential services or physical safety. If you build a regulated or high-impact workflow, provide qualified human review, an appeal path, appropriate notices and controls required by law.
Do not use the service to generate a person’s sensitive profile, infer protected characteristics or make a consequential decision from private data without authorization and safeguards.
4. Security research and testing
You may test your own integration with synthetic or authorized data. Keep probes within your account and the documented API surface. Do not enumerate another customer’s keys, inspect internal routes, attack payment webhooks, bypass authentication, intentionally overload the service or send real customer prompts as shadow traffic.
For a responsible test, use a separate key, a small bounded workload and a clear stop condition. Never include API secrets, personal data or model content in a vulnerability report.
5. Credentials, payments and limits
- Keep keys private, use separate keys per application and revoke exposed credentials immediately.
- Do not share a key, resell access or disguise customer traffic without written permission.
- Do not create accounts, invoices or payment callbacks to evade risk controls, obtain duplicate credits or manipulate ledger state.
- Do not evade request, token, concurrency or spend limits through account rotation, parallel keys, abusive retries or traffic obfuscation.
- Do not use the service to test stolen, leaked, browser-session or third-party credentials.
6. Content and model output
You are responsible for having permission to send input and for reviewing output before publication or action. Model output can be inaccurate, biased, copyrighted, unsafe or incomplete. Do not represent generated content as verified fact without appropriate review, and do not use output to facilitate a prohibited activity.
Do not submit secrets, private keys, passwords, payment credentials or sensitive personal data unless your security, legal and retention controls allow it. Model.sale does not need prompt content to investigate a charge: request ID, status and timestamp are usually sufficient.
7. Reporting abuse or a vulnerability
Report suspected account compromise, payment abuse or a security issue to security@model.sale, or use support for account and billing questions. Include a short description, affected endpoint, approximate time and request ID if available. Do not include API keys, passwords, prompts or response bodies.
If you receive a report involving imminent danger, preserve only the minimum necessary details and route it to the appropriate emergency or legal authority. Model.sale is not an emergency response service.
8. Enforcement and appeals
We may warn, rate-limit, hold a payment, revoke a key, suspend an account, disable a model or terminate access when activity violates this policy, creates a security or payment risk, or is required by law. We may preserve limited security and ledger evidence needed to investigate an incident.
Where appropriate, contact support to explain a false positive or provide remediation details. Do not create a new account to bypass a restriction while an appeal is being reviewed.
9. Updates and contact
We may update this policy as the product, abuse patterns or legal requirements change. The date at the top identifies the current version. Continued use after an update means you accept the revised policy.
General questions: support@model.sale. Security and credential exposure: security@model.sale. Legal notices: legal@model.sale.