deepseek-v4.1POST
15 封真实商务邮件盲测：DeepSeek V4.1 Flash 两次违规，Opus 5.5 零违规
用户用 15 封匿名商务邮件盲测 DeepSeek V4.1 Flash、Opus 5.5 和自己的回复，整体都不错，差距在规则遵守：DeepSeek 用了禁用问候语两次并编造一个承诺，Opus 没违规。
Kevin Theo
@kevintheoo2 天前收录原推发布于 2026.10.26
Kevin Theo @kevintheoo原推发布于 2026.10.26
@EnvolDev different test, same direction. 15 real anonymised business emails, blind: DeepSeek V4.1 Flash vs Opus 5.5 vs my own replies. all pretty good. the only gap was rules: deepseek used a banned greeting twice and invented one promise, opus broke none
x.com/kevintheoo/status/2114440455433973977采集于 2026.10.26 12:03