Kimi K3 open-source multiple core components support large-scale agent training

The Kimi.ai team has opened the Kimi K3 model weights and technical report while gradually open-sourcing multiple underlying infrastructures: including the high-performance communication library MoonEP for distributed MoE training, the distributed environment system AgentENV for large-scale agent workflows (in collaboration with kvcache-ai), and the Kimi Delta Attention high-performance kernel FlashKDA based on CUTLASS.

The official statement claims that these components can reduce communication and inference overhead in large-scale MoE and agent reinforcement learning training, and can be used as a plug-and-play backend for flash-linear-attention.