04:00
2026-07-28
arxiv.org
artificial-intelligence
MegaSlide-DiT: Memory-Centric Adaptation and Deformable Local Attention for Efficient Video Diffusion
Researchers introduce MegaSlide-DiT, a prototype that adapts a pre-trained 105-billion-parameter Diffusion Transformer (DiT) for high-resolution video generation on a single Nvidia H200 GPU with 1.5 Tโฆ