Vectara Hallucination Leaderboard (HHEM)

Measures: The model summarizes source articles using only the text given. Vectara's own hallucination classifier (HHEM-2.3, commercial, not a general LLM judge) flags each summary as consistent or not, with scores below 0.5 counted as hallucinations. The leaderboard reports the share of summaries that hallucinate.

Maintainer: Vectara
Introduced: 2023
Status: Active
Saturation: Near-ceiling
Typical provenance: Independent
In scoring set: Yes

Paper Official leaderboard

Leaderboard

Show independent results only

Every result we hold, whoever ran it. Only independent and third-party rows count towards a score.

No results to show.