Interface AggregateMetrics

Aggregate metrics across a run.

interface AggregateMetrics {
    totalTests: number;
    passedTests: number;
    failedTests: number;
    passRate: number;
    avgScore: number;
    scoreStdDev: number;
    avgLatencyMs: number;
    p50LatencyMs: number;
    p95LatencyMs: number;
    p99LatencyMs: number;
    totalTokens: number;
    totalCostUsd: number;
    byCategory?: Record<string, {
        passRate: number;
        avgScore: number;
        count: number;
    }>;
}

Properties

totalTests: number

Total test cases

passedTests: number

Passed tests

failedTests: number

Failed tests

passRate: number

Pass rate (0-1)

avgScore: number

Average score (0-1)

scoreStdDev: number

Score standard deviation

avgLatencyMs: number

Average latency ms

p50LatencyMs: number

P50 latency

p95LatencyMs: number

P95 latency

p99LatencyMs: number

P99 latency

totalTokens: number

Total tokens used

totalCostUsd: number

Total estimated cost

byCategory?: Record<string, {
    passRate: number;
    avgScore: number;
    count: number;
}>

Metrics by category

Type declaration

  • passRate: number
  • avgScore: number
  • count: number