displayName
MT-Bench
homepageUrl
https://lmsys.org/blog/2023-06-22-leaderboard/
kind
reasoning
targetsKind
ModelVersion
description
MT-Bench (LMSYS) is a multi-turn open-ended question set graded by
a strong-LLM judge to evaluate conversational quality across writing,
reasoning, math, coding, extraction, STEM, and humanities.