Benchmark
MultiChallenge (o3-mini grader)
reasoning
language
text
A realistic multi-turn conversation evaluation benchmark that challenges frontier LLMs across four key areas: instruction retention, inference memory, reliable versioned editing, and self-coherence. Despite near-perfect scores on existing benchmarks, frontier models achieve less than 50% accuracy on MultiChallenge.
Top 10
榜单概览
数据说明
分数为 0-100 百分制综合成绩（Elo 榜为相对评分，仅供同榜内比较）。评测分数属事实数据，整理自公开评测体系与厂商公开资料，部分分数为厂商自报口径，不同榜单的评测方法与数据集可能存在差异，请结合多个榜单交叉参考。
Ranking