Moonshot AI released model weights and technical report for Kimi K3, a 2.8T MoE model with native visual understanding, 1M-token context window, and 2.5x intelligence per unit of compute over prior architecture
Substack is the home for great culture
Moonshot AI released model weights and technical report for Kimi K3, a 2.8T MoE model with native visual understanding, 1M-token context window, and 2.5x intelligence per unit of compute over prior architecture