“Yeah. And so you're also releasing, just to go from the…” — Latent Space: The AI Engineer Podcast at 8:13
“Yeah. And so you're also releasing, just to go from the bottom up, Vidrill, which is a framework for CUDA kernels. So I'm assuming you're going to release all the kernels for like power retention. And then also know you rewrote the flash attention to kernels and actually faster than the ones that 3D originally…”
0:00
32:59