Register and share your invite link to earn from video plays and referrals.

Xiuyu Li
@sheriyuo
Researcher @StepFun_ai | Working on long-horizon tasks | Prev @RUC1937 | Opinions are my own
Joined February 2026
1.9K Following    14.6K Followers
DeepSeek V4.1 Flash 为 552B 参数的 MoE 模型,采用了全新的 Causal-Encoder-Decoder 结构,输入和输出不对称,输入激活只有 8B,输出激活 16B,成本显著低于已知的同尺寸模型。同时,V4.1 Flash 还采用了新的预训练方式、经过了更大规模的强化学习后训练,在基准测试中,成功超越了包括 DeepSeek V4 Pro 在内的一众旗舰模型的智能水平。
Show more
0
107
840
82
Forward to community