# EdgeFusion: On-Device Text-to-Image Generation

arXiv: https://arxiv.org/abs/2404.11925 · HTML: https://mlsanity.com/paper/2404.11925/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models.
- Builds on 22 of its 27 references that are in the map.
- First posted on arXiv 18 Apr 2024.

## Placed in

- Diffusion models / [Diffusion foundations and sampling](https://mlsanity.com/diffusion/diffusion-foundations-and-sampling/)

## Builds on (most cited first, 22 of 22)

- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion](https://mlsanity.com/paper/2208.01618/) (2022)
- [FitNets: Hints for Thin Deep Nets](https://mlsanity.com/paper/1412.6550/) (2014)
- [Latent Consistency Models: Synthesizing High-Resolution Images with Few-Step Inference](https://mlsanity.com/paper/2310.04378/) (2023)
- [Adversarial Diffusion Distillation](https://mlsanity.com/paper/2311.17042/) (2023)
- [On Distillation of Guided Diffusion Models](https://mlsanity.com/paper/2210.03142/) (2022)
- [Scaling up GANs for Text-to-Image Synthesis](https://mlsanity.com/paper/2303.05511/) (2023)
- [Attention is Not All You Need: Pure Attention Loses Rank Doubly Exponentially with Depth](https://mlsanity.com/paper/2103.03404/) (2021)
- [A Comprehensive Overhaul of Feature Distillation](https://mlsanity.com/paper/1904.01866/) (2019)
- [SPHINX: The Joint Mixing of Weights, Tasks, and Visual Embeddings for Multi-modal Large Language Models](https://mlsanity.com/paper/2311.07575/) (2023)
- [SnapFusion: Text-to-Image Diffusion Model on Mobile Devices within Two Seconds](https://mlsanity.com/paper/2306.00980/) (2023)
- [PTQD: Accurate Post-Training Quantization for Diffusion Models](https://mlsanity.com/paper/2305.10657/) (2023)
- [UFOGen: You Forward Once Large Scale Text-to-Image Generation via Diffusion GANs](https://mlsanity.com/paper/2311.09257/) (2023)
- [PIXART-δ: Fast and Controllable Image Generation with Latent Consistency Models](https://mlsanity.com/paper/2401.05252/) (2024)
- [Progressive Knowledge Distillation Of Stable Diffusion XL Using Layer Level Loss](https://mlsanity.com/paper/2401.02677/) (2024)
- [BK-SDM: A Lightweight, Fast, and Cheap Version of Stable Diffusion](https://mlsanity.com/paper/2305.15798/) (2023)
- [Squeezing Large-Scale Diffusion Models for Mobile](https://mlsanity.com/paper/2307.01193/) (2023)

## Built on by (most cited first, 12 of 12)

- [Diffusion Models in Low-Level Vision: A Survey](https://mlsanity.com/paper/2406.11138/) (2024)
- [SnapGen: Taming High-Resolution Text-to-Image Models for Mobile Devices with Efficient Architectures and Training](https://mlsanity.com/paper/2412.09619/) (2024)
- [On-device Sora: Enabling Training-Free Diffusion-based Text-to-Video Generation for Mobile Devices](https://mlsanity.com/paper/2502.04363/) (2025)
- [On-Device Diffusion Transformer Policy for Efficient Robot Manipulation](https://mlsanity.com/paper/2508.00697/) (2025)
- [Individual Content and Motion Dynamics Preserved Pruning for Video Diffusion Models](https://mlsanity.com/paper/2411.18375/) (2024)
- [PocketSR: The Super-Resolution Expert in Your Pocket Mobiles](https://mlsanity.com/paper/2510.03012/) (2025)
- [MoViE: Mobile Diffusion for Video Editing](https://mlsanity.com/paper/2412.06578/) (2024)
- [SnapGen++: Unleashing Diffusion Transformers for Efficient High-Fidelity Image Generation on Edge Devices](https://mlsanity.com/paper/2601.08303/) (2026)
- [Upcycling Text-to-Image Diffusion Models for Multi-Task Capabilities](https://mlsanity.com/paper/2503.11905/) (2025)
- [Towards Redundancy Reduction in Diffusion Models for Efficient Video Super-Resolution](https://mlsanity.com/paper/2509.23980/) (2025)
- [V.I.P. : Iterative Online Preference Distillation for Efficient Video Diffusion Models](https://mlsanity.com/paper/2508.03254/) (2025)
- [On-device Sora: Enabling Training-Free Diffusion-based Text-to-Video Generation for Mobile Devices](https://mlsanity.com/paper/2503.23796/) (2025)

## Abstract (arXiv)

The intensive computational burden of Stable Diffusion (SD) for text-to-image generation poses a significant hurdle for its practical application. To tackle this challenge, recent research focuses on methods to reduce sampling steps, such as Latent Consistency Model (LCM), and on employing architectural optimizations, including pruning and knowledge distillation. Diverging from existing approaches, we uniquely start with a compact SD variant, BK-SDM. We observe that directly applying LCM to BK-SDM with commonly used crawled datasets yields unsatisfactory results. It leads us to develop two strategies: (1) leveraging high-quality image-text pairs from leading generative models and (2) designing an advanced distillation process tailored for LCM. Through our thorough exploration of quantization, profiling, and on-device deployment, we achieve rapid generation of photo-realistic, text-aligned images in just two steps, with latency under one second on resource-limited edge devices.