deepseek-v4.1POST
Rumor: DeepSeek V4.1 Flash is a 763B-parameter model at 200 tok/s
A Chinese post claims DeepSeek released a 763B-parameter Flash model generating around 200 tokens per second, while noting how fast model releases now come.
七月流火
@qiyue_76Archived 9 days agoOriginally posted 2026.10.19
七月流火 @qiyue_76Originally posted 2026.10.19
DeepSeek发布V4.1模型引发实测 深度求索发布参数规模达七百六十三Ｂ的闪电模型 以每秒二百标记的生成速度引发全网实测 现在ai的模型变化是越来越快了啊，真是一月一小变，三月一大变 #V4.1 模型 #DeepSeek #ai模型
x.com/qiyue_76/status/2112003401974497286Archived 2026.10.19 12:05