EpaCache: Error-Propagation-Aware Caching for Accelerating Diffusion-Based Visual Generation

2026-08-29

AI Summary

本文提出EpaCache方法，通过自适应分配重用预算来减少扩散模型生成图像和视频时的推理延迟，同时提高生成质量。

Abstract

Diffusion-based visual generative models deliver strong image and video synthesis quality but incur high inference costs because sequential samplers repeatedly evaluate large networks. Caching-based methods reduce inference latency by reusing intermediate computations across adjacent timesteps. However, existing cache controllers rely primarily on local temporal variation and overlook the trajectory-level consequences of cache reuse. We introduce Error-Propagation-Aware Cache (EpaCache), a training-free caching policy that adaptively allocates the reuse budget on timesteps with lower downstream impact. Experiments on image and video synthesis models demonstrate that EpaCache consistently improves the latency--fidelity trade-off over existing caching methods. On FLUX.1-dev, EpaCache outperforms the prior state-of-the-art caching method in both latency and fidelity, reducing inference time from $11.7$ s to $11.3$ s while improving PSNR from $21.4$ to $22.8$. On HunyuanVideo, EpaCache achieves a $2.63\times$ speedup over uncached inference and improves SSIM from $0.891$ to $0.905$ over the prior state-of-the-art method at matched latency.

Problem

• Diffusion-based Models
• Inference Costs
• Caching Methods
• Temporal Variation

Innovation

• Error-Propagation-Aware
• Caching Policy
• Adaptive Reuse Budget Allocation
• Diffusion Models
• Visual Generation