Type alias LLMMetrics

LLMMetrics: {
    cacheCreationTokens?: number;
    cancelled: boolean;
    completionTokens: number;
    durationMs: number;
    label: string;
    metadata?: MetricsMetadata;
    promptCachedTokens: number;
    promptTokens: number;
    reasoningTokens?: number;
    requestId: string;
    speechId?: string;
    timestamp: number;
    tokensPerSecond: number;
    totalTokens: number;
    ttftMs: number;
    type: "llm_metrics";
}

Type declaration

  • Optional cacheCreationTokens?: number

    Tokens used to write to the prompt cache. Not all providers report this.

  • cancelled: boolean
  • completionTokens: number
  • durationMs: number

    Duration of the request in milliseconds.

  • label: string
  • Optional metadata?: MetricsMetadata

    Metadata for model provider and name tracking.

  • promptCachedTokens: number
  • promptTokens: number
  • Optional reasoningTokens?: number

    The number of completion tokens spent on hidden reasoning.

    Already counted in completionTokens; do not add it to totals. Not all providers break reasoning out separately, and it is 0 when they don't.

  • requestId: string
  • Optional speechId?: string
  • timestamp: number
  • tokensPerSecond: number
  • totalTokens: number
  • ttftMs: number

    Time to first token in milliseconds.

  • type: "llm_metrics"

Settings

Member Visibility

Theme