Pinned Loading

1. Dao-AILab/flash-attention Dao-AILab/flash-attention Public

   Fast and memory-efficient exact attention

   Python 21k 2.2k

2. IDEA-CCNL/Fengshenbang-LM IDEA-CCNL/Fengshenbang-LM Public

   Fengshenbang-LM(封神榜大模型)是IDEA研究院认知计算与自然语言研究中心主导的大模型开源体系，成为中文AIGC和认知智能的基础设施。

   Python 4.1k 381

3. MagiAttention MagiAttention Public

   Forked from SandAI-org/MagiAttention

   A Distributed Attention Towards Linear Scalability for Ultra-Long Context, Heterogeneous Data Training

   Python

4. slime slime Public

   Forked from THUDM/slime

   slime is a LLM post-training framework for RL Scaling.

   Python