Skip to main content

Type Alias: GatewayInferenceCompletionUsage

type GatewayInferenceCompletionUsage = {
[key: string]: unknown;
  completion_tokens: number;
  prompt_tokens: number;
  total_tokens: number;
  completion_tokens_details?:   | {
   [key: string]: unknown;
     reasoning_tokens?: number;
     accepted_prediction_tokens?: number;
     rejected_prediction_tokens?: number;
   }
     | null;
  prompt_tokens_details?:   | {
   [key: string]: unknown;
     cached_tokens?: number;
   }
     | null;
};

Defined in: src/models/ai-gateway-inference.ts:158

Portkey CompletionUsage; additive response fields are preserved.

Example​

`GatewayInferenceCompletionUsageSchema.parse(value);`

Indexable​

[key: string]: unknown

Properties​

completion_tokens​

completion_tokens: number;

Defined in: src/models/ai-gateway-inference.ts:159

prompt_tokens​

prompt_tokens: number;

Defined in: src/models/ai-gateway-inference.ts:160

total_tokens​

total_tokens: number;

Defined in: src/models/ai-gateway-inference.ts:161

completion_tokens_details?​

optional completion_tokens_details?: 
  | {
[key: string]: unknown;
  reasoning_tokens?: number;
  accepted_prediction_tokens?: number;
  rejected_prediction_tokens?: number;
}
  | null;

Defined in: src/models/ai-gateway-inference.ts:162

prompt_tokens_details?​

optional prompt_tokens_details?: 
  | {
[key: string]: unknown;
  cached_tokens?: number;
}
  | null;

Defined in: src/models/ai-gateway-inference.ts:168