Press Space to continue
Mixture-of-Kittens MoE training megakernel reaches 2.37x speedup
Press Space to continue
We're open-sourcing Mixture-of-Kittens (MoK), our MoE training megakernel for NVL72s. It fuses all Mixture-of-Experts communication and computation into a single, fully deterministic kernel, and runs up to 2.37x faster than the strongest public baselines.
open on xViews and clicks count once per visitor per day. Clicks are opens of this post on X.