Schema:

ACCOUNT_USAGE

CORTEX_REST_API_RATE_LIMIT_POLICIES view

Query the CORTEX_REST_API_RATE_LIMIT_POLICIES view to see the per-model rate limit policies for Cortex REST API endpoints in your account. The view shows the requests-per-minute (RPM) and tokens-per-minute (TPM) limits for each model.

Columns

Column NameData TypeDescription
MODEL_NAMETEXTName of the model for which the rate limit policy applies.
RPMNUMBERThe requests-per-minute limit for the model. A NULL value means no RPM limit is enforced.
TPMNUMBERThe tokens-per-minute limit for the model. A NULL value means no TPM limit is enforced.

Usage notes

  • Latency for the view may be up to 360 minutes (6 hours).
  • The view is scoped to the account that runs the query.