# Token Pruning for Caching Better: 9 Times Acceleration on Stable Diffusion for Free

Source: [Token Pruning for Caching Better: 9 Times Acceleration on Stable Diffusion for Free](<https://www.sota2.com/research/paper/68fa766c1e209131cea6ce47>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T03:53:27.652Z

Authors: Evelyn Zhang, Bang Xiao, Jiayi Tang, Qianli Ma, Chang Zou, Xuefei Ning, Xuming Hu, Linfeng Zhang

Published: 2024-12-31T09:56:40.000Z

## Abstract

Stable Diffusion has achieved remarkable success in the field of text-to-image generation, with its powerful generative capabilities and diverse generation results making a lasting impact. However, its iterative denoising introduces high computational costs and slows generation speed, limiting broader adoption. The community has made numerous efforts to reduce this computational burden, with methods like feature caching attracting attention due to their effectiveness and simplicity. Nonetheless, simply reusing features computed at previous timesteps causes the features across adjacent timesteps to become similar, reducing the dynamics of features over time and ultimately compromising the quality of generated images. In this paper, we introduce a dynamics-aware token pruning (DaTo) approach that addresses the limitations of feature caching. DaTo selectively prunes tokens with lower dynamics, allowing only high-dynamic tokens to participate in self-attention layers, thereby extending feature dynamics across timesteps. DaTo combines feature caching with token pruning in a training-free manner, achieving both temporal and token-wise information reuse. Applied to Stable Diffusion on the ImageNet, our approach delivered a 9$\\times$ speedup while reducing FID by 0.33, indicating enhanced image quality. On the COCO-30k, we observed a 7$\\times$ acceleration coupled with a notable FID reduction of 2.17.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2501.00375>)

[PDF](<https://arxiv.org/pdf/2501.00375.pdf>)
