TurboDiffusion accelerates video diffusion models 100-200x on single RTX 5090 GPU, generating 5-second videos in 1.9 seconds using attention acceleration and timestep distillation techniques from Tsinghua researchers

GitHub - thu-ml/TurboDiffusion: TurboDiffusion: 100–200× Acceleration for Video Diffusion Models

TurboDiffusion This repository provides the official implementation of TurboDiffusion, a video generation acceleration framework that can speed up end-to-end diffusion generation by $100 \sim 200\times$ on a single RTX 5090, while maintaining video quality. TurboDiffusion primarily uses SageAttention, SLA (Sparse-Linear Attention) for attention acceleration, and rCM for timestep distillation. Paper: TurboDiffusion: Accelerating Video Diffusion Models by 100-200 Times Note: the checkpoints and pa...