Legal

inference grants you a personal, non-exclusive, non-transferable license to access and use the platform for model routing, API key management, playground testing, and related account workflows. You may not:

• Reverse-engineer, decompile, or extract source code from inference
• Scrape, crawl, or automate requests against the platform without written permission
• Bypass rate limits, credit limits, authentication, billing, or abuse-prevention controls
• Resell, sublicense, or redistribute access to the platform without written permission
• Use inference to send illegal, harmful, infringing, or abusive requests

The materials, APIs, routing behavior, model availability, pricing displays, and platform features on inference are provided on an 'as is' and 'as available' basis.

inference does not warrant that upstream model providers, model outputs, network availability, latency, or pricing will be uninterrupted, error-free, or suitable for every production workload.

You are responsible for activity under your account and API keys. Keep credentials secret, rotate keys when needed, and set limits appropriate for your applications.

You retain rights to prompts, inputs, files, responses, and other content you submit. You grant inference a limited license to transmit, process, cache, store, and display that content only as needed to operate, secure, debug, and improve the service.

You agree not to submit content or traffic that is:

• Illegal, fraudulent, deceptive, or impersonating others
• Infringing on third-party intellectual property or privacy rights
• Spam, phishing, credential harvesting, malware, or automated abuse
• Designed to evade safeguards, rate limits, billing, moderation, or security controls

Paid plans, prepaid credits, and top-ups are used to fund routed model usage and platform access. Credits are not a bank account, stored-value account, or cash equivalent.

Charges may depend on upstream provider pricing, model type, request size, generated output, and platform fees. Displayed estimates are informational and may differ from final settled usage.

Refund requests may be reviewed case by case. Abuse, policy violations, consumed credits, or excessive usage may make a purchase ineligible for refund.

inference may add, remove, rename, rate limit, suspend, or reroute models and endpoints to maintain reliability, comply with provider requirements, address abuse, or improve the platform.

We may prioritize cached status data, temporary fallbacks, or degraded-mode behavior during outages. Model availability and provider behavior can change without notice.

Accounts are for lawful use only. Do not share credentials, evade enforcement, create accounts to bypass limits, or coordinate abuse through multiple accounts or payment methods.

We reserve the right to suspend or terminate accounts, keys, requests, credits, or access at any time for harmful conduct, illegal activity, payment risk, provider enforcement, security risk, or violation of these terms.

To the maximum extent permitted by law, inference, YSN, and their affiliates are not liable for indirect, incidental, consequential, special, exemplary, or punitive damages, including lost profits, lost data, service interruption, or replacement service costs.

• Informal resolution: contact support@inference.ws and allow a 30-day good-faith resolution window before formal action
• Individual disputes: disputes must be brought individually, not as a class action or representative proceeding
• Governing law: legal matters are governed by the jurisdiction where YSN is registered, unless applicable law requires otherwise

We use your information to operate the API router, authenticate users, process requests, meter usage, apply limits, maintain billing records, provide account dashboards, and respond to support or security issues.

We do not sell your personal data to third parties.

Request and usage metadata may be used to detect abuse, debug failures, reconcile billing, monitor reliability, improve routing, and maintain service integrity.

Sensitive credentials and passwords are not stored in plain text. API keys may be hashed or otherwise protected depending on their purpose in the service.

• Account data: retained while your account exists and for a reasonable period after deletion where needed for security, legal, billing, or backup purposes
• Usage and billing records: retained as needed for invoices, credits, audits, abuse prevention, taxes, and legal obligations
• Request content: retained only as needed to provide product features, troubleshoot, secure the service, or comply with applicable obligations
• Security logs: retained for a limited period appropriate to fraud prevention, abuse response, and platform integrity
• Backups: may persist for a limited period after deletion before being overwritten

We use carefully selected providers to deliver the service.

• AI model providers and upstream inference services to process routed requests
• Payment processors for checkout, subscriptions, invoices, and tax or payment records
• Infrastructure, database, storage, analytics, monitoring, and security providers
• Email or support providers when needed for account, product, or support communications

Traffic should be protected in transit with HTTPS/TLS.

Passwords are hashed before storage.

Access controls, sessions, API keys, rate limits, request guards, and monitoring help protect accounts and platform infrastructure.

No internet service can be guaranteed perfectly secure. You are responsible for keeping your account credentials and API keys confidential.

Your data may be processed in countries other than where you live. By using inference, you understand that data may be transferred to infrastructure, providers, and processors used to operate the service.