SnapFusion: Text-to-Image Diffusion Model on Mobile Devices within Two Seconds
This work democratizes content creation by making powerful text-to-image diffusion models accessible on mobile devices, addressing cost and privacy issues associated with cloud-based inference.
The paper tackles the computational inefficiency of text-to-image diffusion models by introducing an efficient network architecture and improved step distillation, enabling these models to run on mobile devices in under 2 seconds while achieving better FID and CLIP scores than Stable Diffusion v1.5 with fewer steps.
Text-to-image diffusion models can create stunning images from natural language descriptions that rival the work of professional artists and photographers. However, these models are large, with complex network architectures and tens of denoising iterations, making them computationally expensive and slow to run. As a result, high-end GPUs and cloud-based inference are required to run diffusion models at scale. This is costly and has privacy implications, especially when user data is sent to a third party. To overcome these challenges, we present a generic approach that, for the first time, unlocks running text-to-image diffusion models on mobile devices in less than $2$ seconds. We achieve so by introducing efficient network architecture and improving step distillation. Specifically, we propose an efficient UNet by identifying the redundancy of the original model and reducing the computation of the image decoder via data distillation. Further, we enhance the step distillation by exploring training strategies and introducing regularization from classifier-free guidance. Our extensive experiments on MS-COCO show that our model with $8$ denoising steps achieves better FID and CLIP scores than Stable Diffusion v$1.5$ with $50$ steps. Our work democratizes content creation by bringing powerful text-to-image diffusion models to the hands of users.