跳到正文
原文
论文Artificial Analysis · X· @ArtificialAnlys·· 1 小时前

3593 次浏览 · 16 赞 · 2 转发 (收录时)

在以问句结尾的回复比例测量中,MiMo-V2.6-Pro 位居榜首

谁
Artificial Analysis
引文
In AA-PersonalityBench, MiMo-V2.6-Pro ends 34% of its responses with a question, the highest share of any model measured

发生了什么

在 AA-PersonalityBench 中,MiMo-V2.6-Pro 有 34% 的回复以问句结尾,在所有受测模型中比例最高 在受测模型中,以问句结尾的回复比例从 34% 到 0% 不等。有六个模型以 结尾的回复不到 0.5%

原帖含 1 个链接

接下来读什么

查看这件事的完整进展Artificial Analysis Finds MiMo-V2.6-Pro Ends 34% of Responses With Questions
正文 · 原文(语言尚未确认)

In AA-PersonalityBench, MiMo-V2.6-Pro ends 34% of its responses with a question, the highest share of any model measured

The share of responses that end with a question ranges from 34% down to 0% across the models measured. Six models end fewer than 0.5% of their responses with https://t.co/VnRVlo7b6g

这件事到目前为止

Artificial Analysis reports that MiMo-V2.6-Pro ended 34% of its responses with a question in AA-PersonalityBench, the highest share among the models measured. Across those models, the rate ranged from 34% to 0%; six ended fewer than 0.5% of their responses with a question.

每天一封,跟上 AI 的重要变化

模型发布、价格变化、算力进展,附原始来源。免费,确认后开始发送,随时一键退订。 看一份往期日报 · 2026-11-05