Skip to main content

Usage Reports

Management > Usage records your service and resource usage on the platform in three reports, each queryable for the past 7, 14, or 30 days, or a custom range (up to 30 days).

Model Inference Usage​

A record of every model call:

  • Date and time, status (whether the operation succeeded)
  • Model type (Chat / Image Generation / Audio Translation / Audio Transcription / Audio Speech / Embedding, and so on, depending on what is published on the platform)
  • Endpoint/model, total amount
  • Usage: the billing unit varies by model type

Pods Usage​

Resource usage records for every Pod and private endpoint:

  • Pod name, type (Pod / Endpoint), resource specification
  • Running status (Pending / Running / Terminated), usage duration
  • Rate (billed hourly), total amount
  • Creation time, deletion time, instance ID

Click to expand a record and view the start/end times, replica count, usage duration, and amount breakdown.

Storage Usage​

Usage records for every volume:

  • Volume name, resource specification (capacity), status
  • Usage duration, rate (billed hourly), total amount
  • Creation time, deletion time, volume ID

Click to expand a record and view the start/end times, capacity, usage duration, and amount breakdown.