Evaluation
Hallucination
A hallucination is a confident, fluent output that is factually wrong or unsupported.
The model completes a plausible-sounding answer rather than admitting that it does not know.
It is a structural property of how these models generate text, not a bug to patch out.
Grounding through RAG and evals, not a better prompt, are the actual mitigations.
A demo that never shows a hallucination usually means its test set was too easy.
← Back to the full glossaryRelated terms
Retrieval-Augmented Generation (RAG) Retrieval-Augmented Generation (RAG) retrieves relevant passages at query time and feeds them into an LLM's context. Prompt Injection Prompt injection is untrusted input crafted to override a model's system prompt or task. Eval Suite (Evals) An eval suite is a versioned set of test cases that measures whether outputs are good.
More in Evaluation
LLM-as-Judge LLM-as-judge uses another LLM to score a model's outputs against a rubric, at scale. Observability Observability captures traces of every prompt, retrieval, tool call, and response for debugging. Shadow Deployment A shadow deployment runs a candidate release against real production traffic, offline, before release.