HellaSwag
更新于 2026-04-03
1,030 次浏览
问题数量
70000
发布机构
University of Washington
评测类别
常识推理
评测指标
Accuracy
支持语言
英语
难度等级
Intermediate
简介
一个包含 70,000 个多项选择题的基准，用于评估模型的常识推理能力。
相关资源
HellaSwag Model Score Leaderboard
Source: DataLearnerAI
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
模式说明:
normal
thinking
low
medium
high
deeper thinking
parallel_thinking
图表加载中...
Latest HellaSwag model rankings and full benchmark leaderboard
Browse the latest scores, model modes, release dates, and parameter sizes for HellaSwag.
Model release cutoff:
HellaSwag详细排名数据表格
| 排名 | 模型 | |||
|---|---|---|---|---|
| 1 | Claude3-Opus Standard Mode | 95.40 | 2024-03-04 | Unknown |
| 2 | Gemma2-27B Standard Mode | 86.40 | 2024-05-14 | 27B |
| 3 | Gemma 3 - 27B (IT) Standard Mode | 85.60 | 2025-03-12 | 27B |