FasterCache: Training-Free Video Diffusion Model Acceleration with High Quality

TL;DR

FasterCache is a training-free method that accelerates video diffusion model inference by intelligently reusing features and optimizing classifier-free guidance, achieving significant speedups without quality loss.

Contribution

The paper introduces FasterCache, a novel training-free approach that enhances inference speed of video diffusion models through dynamic feature reuse and CFG optimization.

Findings

Achieves 1.67× speedup on Vchitect-2.0

Maintains comparable video quality to baseline

Outperforms existing acceleration methods

Abstract

In this paper, we present \textbf{\textit{FasterCache}}, a novel training-free strategy designed to accelerate the inference of video diffusion models with high-quality generation. By analyzing existing cache-based methods, we observe that \textit{directly reusing adjacent-step features degrades video quality due to the loss of subtle variations}. We further perform a pioneering investigation of the acceleration potential of classifier-free guidance (CFG) and reveal significant redundancy between conditional and unconditional features within the same timestep. Capitalizing on these observations, we introduce FasterCache to substantially accelerate diffusion-based video generation. Our key contributions include a dynamic feature reuse strategy that preserves both feature distinction and temporal continuity, and CFG-Cache which optimizes the reuse of conditional and unconditional outputs to further enhance inference speed without compromising video quality. We empirically evaluate FasterCache on recent video diffusion models. Experimental results show that FasterCache can significantly accelerate video generation (\eg 1.67$\times$ speedup on Vchitect-2.0) while keeping video quality comparable to the baseline, and consistently outperform existing methods in both inference speed and video quality.