Kimi-K2 Collection Moonshot's MoE LLMs with 1 trillion parameters, exceptional on agentic intellegence • 5 items • Updated 27 days ago • 159
Kimi Linear: An Expressive, Efficient Attention Architecture Paper • 2510.26692 • Published Oct 30 • 117 • 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 Oct 30 • 117
Moonlight-A3B Collection Moonshot's Compute-efficient MoE LLM, first Scaling Up of Muon Optimizer • 3 items • Updated Nov 2 • 8