AA-PersonalityBench：MiMo-V2.6-Pro 34% 回复以提问收尾居首
看看谁最爱反问：MiMo-V2.6-Pro 三成回复结尾抛问题，DeepSeek V4.1 Flash 几乎从不，对比很直观。
Artificial Analysis 在 AA-PersonalityBench 中统计各模型以问句结尾回复的比例，MiMo-V2.6-Pro 达 34%，为所有受测模型最高，GPT-6 Astra 以 32%、GPT-6.1 Sol 以 29% 紧随其后。最低一侧，GPT-5.4 mini 和 Nemotron 3 Ultra 均为 0.0%，DeepSeek V4.1 Flash 为 0.1%，GPT-3.5 Turbo 为 0.2%，Grok 4.7 为 0.4%。六个模型的该比例低于 0.5%，模型间反问倾向差异跨度从 34% 到 0%。
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 a question.
Share of responses that end with a question:
➤ Most often: MiMo-V2.6-Pro (34%), GPT-6 Astra (32%), GPT-6.1 Sol (29%), Muse Glimmer (28%) and Muse Spark Open (26%)
➤ Least often: GPT-5.4 mini (0.0%), Nemotron 3 Ultra (0.0%), DeepSeek V4.1 Flash (0.1%), DeepSeek V4.1 Flash 1017 (0.1%), GPT-3.5 Turbo (0.2%) and Grok 4.7 (0.4%)
