SandAI-org/MagiAttention A Distributed Attention Towards Linear Scalability for Ultra-Long Context, Heterogeneous Data Training **Language:**Python Total stars: 86 Stars trend: 21 Apr 2025 11am ▍ +3 12pm ██▉ +23 1pm ██▉ +23 2pm █▍ +11 3pm █▏ +9 4pm ▉ +7 #python