Clockwork Diffusion: Efficient Generation With Model-Step Distillation
This addresses the computational bottleneck in diffusion models for users needing faster image generation, though it is incremental as it builds on existing methods.
This work tackles the inefficiency of text-to-image diffusion models by identifying that not all UNet operations are equally important, and proposes Clockwork Diffusion to reuse computations from previous steps, saving 32% of FLOPs with negligible quality loss in Stable Diffusion v1.5.
This work aims to improve the efficiency of text-to-image diffusion models. While diffusion models use computationally expensive UNet-based denoising operations in every generation step, we identify that not all operations are equally relevant for the final output quality. In particular, we observe that UNet layers operating on high-res feature maps are relatively sensitive to small perturbations. In contrast, low-res feature maps influence the semantic layout of the final image and can often be perturbed with no noticeable change in the output. Based on this observation, we propose Clockwork Diffusion, a method that periodically reuses computation from preceding denoising steps to approximate low-res feature maps at one or more subsequent steps. For multiple baselines, and for both text-to-image generation and image editing, we demonstrate that Clockwork leads to comparable or improved perceptual scores with drastically reduced computational complexity. As an example, for Stable Diffusion v1.5 with 8 DPM++ steps we save 32% of FLOPs with negligible FID and CLIP change.