# MobileDiffusion: Subsecond Text-to-Image Generation on Mobile Devices

**Authors:** Yang Zhao, Yanwu Xu, Zhisheng Xiao, Tingbo Hou  
**arXiv:** [2311.16567](https://arxiv.org/abs/2311.16567)  
**Date:** 2023-11-28  
**Summary page:** https://www.summarizepaper.com/fr/paper/2311.16567v1/

## Abstract

The deployment of large-scale text-to-image diffusion models on mobile devices is impeded by their substantial model size and slow inference speed. In this paper, we propose \textbf{MobileDiffusion}, a highly efficient text-to-image diffusion model obtained through extensive optimizations in both architecture and sampling techniques. We conduct a comprehensive examination of model architecture design to reduce redundancy, enhance computational efficiency, and minimize model's parameter count, while preserving image generation quality. Additionally, we employ distillation and diffusion-GAN finetuning techniques on MobileDiffusion to achieve 8-step and 1-step inference respectively. Empirical studies, conducted both quantitatively and qualitatively, demonstrate the effectiveness of our proposed techniques. MobileDiffusion achieves a remarkable \textbf{sub-second} inference speed for generating a $512\times512$ image on mobile devices, establishing a new state of the art.