Communication-Efficient Diffusion Denoising Parallelization via Reuse-then-Predict Mechanism
This addresses the deployment bottleneck for diffusion models in bandwidth-constrained environments, offering a scalable solution with incremental improvements over existing parallelization strategies.
The paper tackles the high inference latency of diffusion models by proposing ParaStep, a parallelization method that reduces communication overhead through a reuse-then-predict mechanism, achieving speedups of up to 3.88x on SVD, 2.43x on CogVideoX-2b, and 6.56x on AudioLDM2-large while maintaining quality.
Diffusion models have emerged as a powerful class of generative models across various modalities, including image, video, and audio synthesis. However, their deployment is often limited by significant inference latency, primarily due to the inherently sequential nature of the denoising process. While existing parallelization strategies attempt to accelerate inference by distributing computation across multiple devices, they typically incur high communication overhead, hindering deployment on commercial hardware. To address this challenge, we propose \textbf{ParaStep}, a novel parallelization method based on a reuse-then-predict mechanism that parallelizes diffusion inference by exploiting similarity between adjacent denoising steps. Unlike prior approaches that rely on layer-wise or stage-wise communication, ParaStep employs lightweight, step-wise communication, substantially reducing overhead. ParaStep achieves end-to-end speedups of up to \textbf{3.88}$\times$ on SVD, \textbf{2.43}$\times$ on CogVideoX-2b, and \textbf{6.56}$\times$ on AudioLDM2-large, while maintaining generation quality. These results highlight ParaStep as a scalable and communication-efficient solution for accelerating diffusion inference, particularly in bandwidth-constrained environments.