Skip to Content
Anuma SDKClientInternalType AliasesOpenaiCompletionUsageCompletionTokensDetails

OpenaiCompletionUsageCompletionTokensDetails

OpenaiCompletionUsageCompletionTokensDetails = object

Defined in: src/client/types.gen.ts:3737 

Breakdown of tokens used in a completion.

Properties

accepted_prediction_tokens?

optional accepted_prediction_tokens: number

Defined in: src/client/types.gen.ts:3742 

When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion.


audio_tokens?

optional audio_tokens: number

Defined in: src/client/types.gen.ts:3746 

Audio input tokens generated by the model.


reasoning_tokens?

optional reasoning_tokens: number

Defined in: src/client/types.gen.ts:3750 

Tokens generated by the model for reasoning.


rejected_prediction_tokens?

optional rejected_prediction_tokens: number

Defined in: src/client/types.gen.ts:3757 

When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits.

Last updated on