Single-agent Baseline
A narrowly instructed model with approved context and a small tool set, used as the measurable default before adding more loops or agents.
The single-agent baseline is the simplest credible design: one bounded outcome, relevant retrieval or context, a short tool list and clear success examples. It is the control condition against which added planners, workers and evaluators must earn their cost.
Use it for knowledge questions, extraction, classification, summarisation and one-step actions. Keep permissions, data access, validation and irreversible changes in deterministic code around it.
Stadig nysgerrig?
Bed Kuni, din AI-læringspartner, om at forklare dette koncept i konteksten af dit eget arbejde.
AI kan lave fejl. Kontrollér vigtige fakta, beslutninger og kilder, før du stoler på dem.
Spørg MAESTRO
Spørg PRISM
Spørg KuniSTART
Bounded roles
CONTROL
Visible hand-offs
OUTCOME
Human owner
AGENTS & WORKFLOWS
Gå dybere
Relaterede koncepter
Set i cases
Eksempler fra virkeligheden, hvor dette koncept optræder i vores cases.
Klar til at bringe AI ind i jeres organisation?
Tal med os om en guidet adoptionsvej for jeres team — fra første use case til produktion.
Spørg om dette koncept