Skip to main content

Inference and models FAQ

No. The serverless inference API and the playground have been retired. The models that used to be listed are no longer available, which is why they do not appear in the app.We are building a new inference product. We cannot share a release date yet.
Rent an on-demand GPU and run an open-source model with your own serving stack.To tell us what you need from the new product, email support@hyperbolic.ai with the models you used, roughly how many requests you send, and what you use them for.
Requests to the old inference models no longer work, even with a valid API key. Code that calls the old inference endpoints will keep failing until the new product launches.
Credit can still be used for GPU rentals and storage. See Refunds and withdrawals if you want to ask about unused credit.