When an LLM confidently states facts that are false or fabricates citations. The gap between output plausibility and output correctness.
reliability
Hallucination
Related terms
Tools that implement Hallucination
LA
LangSmith
Evaluation · Platform (any LLM)
8.7
LangChain's eval + observability platform.
Freemium· Developer: $0 · Plus: $39 · Enterprise: Custom pricingLLM tracingevals
BR
Braintrust
FeaturedEvaluation · Platform (any LLM)
8.9
Eval, monitor, and improve AI products end-to-end.
Freemium· Starter: $0 · Pro: $249 · Enterprise: Custom pricingevalsmonitoring