avgLatencyBaseline
avgLatencyBaseline: number
avgLatencyCandidate
avgLatencyCandidate: number
avgTokenBaseline
avgTokenBaseline: number
avgTokenCandidate
avgTokenCandidate: number
avgToolCallBaseline: number
avgToolCallCandidate: number
baselinePassRate
baselinePassRate: number
baselineSpecId
baselineSpecId: string
candidatePassRate
candidatePassRate: number
candidateSpecId
candidateSpecId: string
choiceAwarenessBaselineRate
choiceAwarenessBaselineRate: number
choiceAwarenessCandidateRate
choiceAwarenessCandidateRate: number
deterministicSeed
deterministicSeed: number
experimentId
experimentId: string
overallScoreCi95
overallScoreCi95: [number, number]
overallScoreEffectSize
overallScoreEffectSize: number
passRateUplift
passRateUplift: number
recommendations
recommendations: string[]
reportId
reportId: string
reportSha256
reportSha256: string
totalProbes
totalProbes: number
Aggregate metrics