跳到正文
RCreddit.com·
暂不在当前实时榜单

Humor Arena: Which LLM is the funniest?

AI 摘要

In a humor arena comparing 20 LLM versions on 360 joke prompts, Fable 5 achieved the highest estimated score of 66.8 points per 100 comparisons. Fable 5.1 followed with 58.2 points. A humor-trained judge evaluated four jokes per prompt, with model names hidden. Scores were calculated by assigning one point for a win and half a point for a tie, and models near the top had overlapping uncertainty intervals.

为什么是这条

This report uniquely details a humor-focused LLM comparison, unlike typical benchmarks, using 360 frozen joke prompts and a humor-trained judge to rank 20 model versions.

时间与来源

时间显示为 UTC

显示时区:UTC

本地时区尚不可用,暂时显示 UTC。

收录当时偏移:UTC+02026年9月18日 07:00 UTC

收录
2026年9月18日 07:00
来源类型
开发者社区

讨论趋势

→ 平稳
最近 24 小时与此前 24 小时的快照均值对比 · 7 天曲线

百分比基于采集到的讨论信号,不代表新增评论数或独立参与人数。曲线仅用于同一话题在不同时段的比较。

本站未收录正文。

前往源站阅读 →
来源·reddit.com