Efficient-vDiT: Efficient Video Diffusion Transformers With Attention Tile

TL;DR

Efficient-vDiT introduces a sparse attention mechanism and multi-step distillation to significantly accelerate video diffusion transformer inference, reducing computation time with minimal quality loss.

Contribution

This paper proposes a novel sparse 3D attention method and a multi-step distillation approach to improve the efficiency of video diffusion transformers.

Findings

Achieves 7.4x-7.8x faster inference on 29 and 93 frame videos.

Reduces inference time by leveraging sparse attention and distillation.

Maintains comparable video quality with minimal performance trade-offs.

Abstract

Despite the promise of synthesizing high-fidelity videos, Diffusion Transformers (DiTs) with 3D full attention suffer from expensive inference due to the complexity of attention computation and numerous sampling steps. For example, the popular Open-Sora-Plan model consumes more than 9 minutes for generating a single video of 29 frames. This paper addresses the inefficiency issue from two aspects: 1) Prune the 3D full attention based on the redundancy within video data; We identify a prevalent tile-style repetitive pattern in the 3D attention maps for video data, and advocate a new family of sparse 3D attention that holds a linear complexity w.r.t. the number of video frames. 2) Shorten the sampling process by adopting existing multi-step consistency distillation; We split the entire sampling trajectory into several segments and perform consistency distillation within each one to activate few-step generation capacities. We further devise a three-stage training pipeline to conjoin the low-complexity attention and few-step generation capacities. Notably, with 0.1% pretraining data, we turn the Open-Sora-Plan-1.2 model into an efficient one that is 7.4x -7.8x faster for 29 and 93 frames 720p video generation with a marginal performance trade-off in VBench. In addition, we demonstrate that our approach is amenable to distributed inference, achieving an additional 3.91x speedup when running on 4 GPUs with sequence parallelism.