Skip to main content

Usage, Billing and Provisioned Throughput

You can view usage and billing information for Managed Inference model consumption through dedicated views on the Intelligence Foundry. Steps are listed below:

In order to view usage and billing information for Managed Inference model consumption via the Intelligence Foundry console:

  • Login to the Crusoe Cloud console
  • Navigate to the Intelligence Foundry using the switcher in the top right corner
  • Click on the 'Usage' tab in the left nav to view usage information broken up by models and token types
  • Click on the 'Billing' tab in the left nav to view billing information

Provisioned Throughput​

For production applications that require consistent performance, we recommend using the provisioned throughput model. This model guarantees a minimum level of performance for your model by guaranteeing a minimum token per minute target. Please reach out to Crusoe Cloud support if interested in enabling this for your models.