Inference APIs

Terms of Service

Last updated: September 17, 2026. This replaces the terms of March 20, 2025.

These terms govern your use of the website inferenceapis.com and the API at api.inferenceapis.com (together, the "service"), operated by Inference APIs ("we", "us"). By creating an account or using the service you agree to them.

1. The service

We provide hosted access to third-party open-weight AI models through an OpenAI-compatible API, with browser playgrounds and documentation. Models are run on infrastructure operated by our upstream provider. We may add, change or withdraw models; when a model you can call is going to be withdrawn we will say so in the changelog with as much notice as we are given.

2. Accounts and API keys

  • You must give accurate registration details and be old enough to form a contract where you live.
  • You are responsible for everything done with your account and your API keys. Keep keys secret; if one leaks, delete it on the API Keys page and create a new one. Usage made with a leaked key before you delete it is charged to your balance.

3. Prepaid credits and pricing

  • The service is paid for from a prepaid balance in US dollars. Payments are processed by Stripe.
  • Each successful request is charged at the rate published for its model on the pricing page at the time of the request: per token, per character or per audio minute. Requests that fail with a client or server error are not charged.
  • When your balance reaches zero, requests are refused with HTTP 402 until you add credit. We do not bill in arrears and there are no subscriptions or minimums.
  • Prices can change as our own costs change. Changes are published on the pricing page and in the changelog and apply to requests made after the change. Credit already purchased keeps its dollar value.
  • Purchased credit is non-refundable except where the law requires otherwise. If you were charged for something that went wrong on our side, write to us and we will put it right.

4. Acceptable use

You may not use the service to:

  • break the law, or infringe or misappropriate anyone's rights;
  • generate sexual content involving minors, or content that promotes violence against or harassment of individuals or groups;
  • deceive people about the fact that they are interacting with an AI where that matters to them, impersonate real people without consent using synthetic speech, or produce fraud, spam or malware;
  • make fully automated decisions with legal or similarly significant effects on people without appropriate human review;
  • attack, overload, probe or circumvent the security or metering of the service, or resell access in a way that hides abusive traffic from us;
  • violate the licence or acceptable-use policy of the model you call. Each model page names the model; its licence is published by its author.

We may throttle, suspend or close accounts that do these things. Where we reasonably can, we will contact you first.

5. Your content and the output

You keep all rights you have in what you send to the API. As between you and us, you own the output, to the extent the law allows output to be owned. We do not store request or response content and do not use it to train models; see the privacy policy. Models make mistakes: output can be inaccurate, biased or unsafe, similar requests from different users can produce similar output, and you are responsible for checking output before relying on it.

6. Availability

We work to keep the service available and publish its state on the status page, but it is provided without a service-level agreement. Models depend on upstream capacity and can be temporarily unavailable. Build retries and, for anything important, a fallback.

7. Our content

The website, documentation and reference material are ours. You may quote and link to them with attribution. Code samples on the site may be used freely in your own projects.

8. Disclaimers and liability

The service is provided "as is" and "as available", without warranties of any kind, to the fullest extent the law allows. We are not liable for indirect, incidental, special or consequential damages, or for lost profits, revenue or data. Our total liability for any claim relating to the service is limited to the amount you paid us in the three months before the event giving rise to the claim. Nothing in these terms limits liability that cannot be limited by law.

9. Ending the agreement

You can stop using the service and ask us to delete your account at any time. We can suspend or end your access for breach of these terms, for non-trivial risk to the service or to others, or if we stop offering the service; if we close the service for reasons other than your breach, we will refund unused purchased credit.

10. Changes

We may update these terms. The date at the top will change, material changes will be noted in the changelog and emailed to account holders, and continued use after a change means you accept it.

11. Governing law

These terms are governed by the laws of the United States and the state in which Inference APIs is established, without regard to conflict-of-law rules. Disputes will be resolved in the courts located there, unless the law where you live gives you a right to bring them elsewhere.

12. Contact

Inference APIs · support@inferenceapis.com · contact page