Kimi-K2 Collection Moonshot's MoE LLMs with 1 trillion parameters, exceptional on agentic intellegence • 5 items • Updated 15 days ago • 154
Kimi Linear: An Expressive, Efficient Attention Architecture Paper • 2510.26692 • Published 30 days ago • 112 • 4
view article Article Ring-flash-linear-2.0: A Highly Efficient Hybrid Architecture for Test-Time Scaling Oct 9 • 11
Kimi Linear: An Expressive, Efficient Attention Architecture Paper • 2510.26692 • Published 30 days ago • 112
Moonlight-A3B Collection Moonshot's Compute-efficient MoE LLM, first Scaling Up of Muon Optimizer • 3 items • Updated 27 days ago • 8