Skip to content

How are inference requests billed and limited?

From the EverXCloud help center.

When Token market is shown in your portal (/inference), use it as described below. If you do not see it, it is not open for your account yet. When available, text inference bills measured input and output tokens; image models have per-image pricing shown in Token market (/inference). A request reserves its maximum cost, including max_tokens, against your available balance. Unused funds are released when the request ends. Insufficient funds can reject a call before generation. Request-rate and generation-time limits also apply. Check the selected model and recent requests in the portal rather than assuming a fixed price or an unlimited quota.

Contact Support

Submit a request