Skip to main content

COMPEL Glossary / sli-slo-for-ai

SLI/SLO for AI

Service-level indicators and objectives for AI systems — including evaluation score, per-task cost, and goal-achievement rate alongside classical availability/latency.

What this means in practice

Unlike traditional SLOs, AI SLOs must account for non-deterministic output quality and drift; their design is itself an architectural artefact.

Synonyms

AI SLI , AI SLO , GenAI service-level objectives

See also

  • Observability for AI — The set of telemetry — prompt/response capture, retrieval traces, tool-call records, token-cost metrics, and evaluation signals — that makes an AI system operable and auditable.
  • TTFT (time-to-first-token) — The latency from request submission to the first streamed output token.
  • Goal-achievement rate — An agent service-level indicator measuring the fraction of tasks that reach their intended outcome without human intervention.
  • Per-task cost — An agent SLI capturing the full compute and API cost of a single task end-to-end — including all loop iterations, tool calls, memory reads and writes.