EpaCache智能分配缓存，让生成图像视频更快更准。
EpaCache: Error-Propagation-Aware Caching for Accelerating Diffusion-Based Visual Generation
- 根据误差传播影响动态分配缓存资源，避免关键步骤丢精度。
- 在FLUX.1-dev上提速4%（11.7→11.3秒），PSNR提升至22.8。
- 适用于追求高效率与高质量的扩散模型推理部署者。
基于扩散的视觉生成模型虽能产出高质量图像与视频，但因序列采样器反复调用大型网络而带来高昂推理成本。缓存方法通过复用相邻时间步的中间计算来降低延迟，但现有缓存控制器主要依赖局部时序变化，忽视缓存复用对整体轨迹的影响。本文提出无需训练的误差传播感知缓存策略EpaCache，自适应地将复用预算分配给下游影响较小的时间步。在图像与视频生成模型上的实验表明，EpaCache持续优于现有缓存方法。在FLUX.1-dev上，相较于之前最优缓存方法，推理时间从11.7秒降至11.3秒，PSNR由21.4提升至22.8；在HunyuanVideo上，相比未缓存推理实现2.63倍加速，且在相同延迟下SSIM从0.891提升至0.905。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。