Interface CompletionTokensDetails

Breakdown of tokens used in a completion.

interface CompletionTokensDetails {
    accepted_prediction_tokens?: number;
    audio_tokens?: number;
    reasoning_tokens?: number;
    rejected_prediction_tokens?: number;
}

Properties

accepted_prediction_tokens? audio_tokens? reasoning_tokens? rejected_prediction_tokens?

Properties

Optionalaccepted_prediction_tokens

accepted_prediction_tokens?: number

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

Optionalaudio_tokens

audio_tokens?: number

Audio input tokens generated by the model.

Optionalreasoning_tokens

reasoning_tokens?: number

Tokens generated by the model for reasoning.

Optionalrejected_prediction_tokens

rejected_prediction_tokens?: number

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.

Settings

Member Visibility

On This Page

Properties