def log_endpoint_metric(
    endpoint_name: str,
    metric_name: str,
    value: float,
    measured_at: Optional[Union[str, datetime.datetime]] = None,
    api_key: Optional[str] = None,
) -> Dict[str, str]:

source code

Append speed or cost data to the standardized time-series benchmarks for a custom endpoint (only custom endpoints are publishable by end users).

Arguments:

  • endpoint_name - Name of the custom endpoint to append benchmark data for.
  • metric_name - Name of the metric to submit. Allowed metrics are: “input_cost”,
  • value - Value of the metric to submit.
  • measured_at - The timestamp to associate with the submission. Defaults to current
  • api_key - If specified, unify API key to be used. Defaults to the value in the
get_endpoint_metricsContext