CodePipelineTokenUsageEntry

A count of one type of token a step used, with the model and credential that used them.

Kind: an object, used in responses.

Fields

Every field is present in a response; empty ones are null.

Field	Type	Description
type	string	The token type, as the provider reports it, for example input, output, cache_read, cache_creation_5m or cache_creation_1h.
amount	integer	How many tokens of this type.
model	string | null	The model the tokens are counted against: wireModel when the coding tool reported one, otherwise resolvedModel. Null when unknown.
wireModel	string | null	The model the coding tool reported it actually ran on. Null when the coding tool doesn’t report one.
requestedModel	string | null	The model the run’s agentOverrides asked for, for this step’s agent. Null when no override named one.
resolvedModel	string | null	The model chosen for this step’s agent: the override’s model, else the agent’s defaultModel.
credentialId	string | null	The credentialId of the credential that used these tokens. Null when unknown.
tier	string | null	The model’s pricing tier these tokens fell in, for models priced by request size (for example large-context). Null: the model’s standard rate.

Example

{
  "type": "...",
  "amount": 0,
  "model": "...",
  "wireModel": "...",
  "requestedModel": "...",
  "resolvedModel": "...",
  "credentialId": "...",
  "tier": "..."
}

Where it’s used

• Inside CodePipelineStepExecution