scoring
The multi-model evaluator that compares scoring outputs across LLM providers to keep results consistent.
Consensus Engine is an internal safety net: every score is recomputed across multiple LLM providers, and divergent results either get flagged for review or settle on a moderated value. The mechanism keeps the viability score stable across vendor drift.
Source: curated glossary catalog. Freshness: git_versioned_curated_catalog.
Term API · apps/web/data/glossary/terms.ts