DreamLite: A Lightweight On-Device Unified Model for Image Generation and Editing

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Feng, Kailai, Wei, Yuxiang, Chen, Bo, Pan, Yang, Ye, Hu, Liu, Songwei, Yan, Chenqian, Gao, Yuan
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914431923912704
author Feng, Kailai
Wei, Yuxiang
Chen, Bo
Pan, Yang
Ye, Hu
Liu, Songwei
Yan, Chenqian
Gao, Yuan
author_facet Feng, Kailai
Wei, Yuxiang
Chen, Bo
Pan, Yang
Ye, Hu
Liu, Songwei
Yan, Chenqian
Gao, Yuan
contents Diffusion models have made significant progress in both text-to-image (T2I) generation and text-guided image editing. However, these models are typically built with billions of parameters, leading to high latency and increased deployment challenges. While on-device diffusion models improve efficiency, they largely focus on T2I generation and lack support for image editing. In this paper, we propose DreamLite, a compact unified on-device diffusion model (0.39B) that supports both T2I generation and text-guided image editing within a single network. DreamLite is built on a pruned mobile U-Net backbone and unifies conditioning through in-context spatial concatenation in the latent space. It concatenates images horizontally as input, using a (target | blank) configuration for generation tasks and (target | source) for editing tasks. To stabilize the training of this compact model, we introduce a task-progressive joint pretraining strategy that sequentially targets T2I, editing, and joint tasks. After high-quality SFT and reinforcement learning, DreamLite achieves GenEval (0.72) for image generation and ImgEdit (4.11) for image editing, outperforming existing on-device models and remaining competitive with several server-side models. By employing step distillation, we further reduce denoising processing to just 4 steps, enabling our DreamLite could generate or edit a 1024 x 1024 image in less than 1s on a Xiaomi 14 smartphone. To the best of our knowledge, DreamLite is the first unified on-device diffusion model that supports both image generation and image editing.
format Preprint
id arxiv_https___arxiv_org_abs_2603_28713
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle DreamLite: A Lightweight On-Device Unified Model for Image Generation and Editing
Feng, Kailai
Wei, Yuxiang
Chen, Bo
Pan, Yang
Ye, Hu
Liu, Songwei
Yan, Chenqian
Gao, Yuan
Computer Vision and Pattern Recognition
Diffusion models have made significant progress in both text-to-image (T2I) generation and text-guided image editing. However, these models are typically built with billions of parameters, leading to high latency and increased deployment challenges. While on-device diffusion models improve efficiency, they largely focus on T2I generation and lack support for image editing. In this paper, we propose DreamLite, a compact unified on-device diffusion model (0.39B) that supports both T2I generation and text-guided image editing within a single network. DreamLite is built on a pruned mobile U-Net backbone and unifies conditioning through in-context spatial concatenation in the latent space. It concatenates images horizontally as input, using a (target | blank) configuration for generation tasks and (target | source) for editing tasks. To stabilize the training of this compact model, we introduce a task-progressive joint pretraining strategy that sequentially targets T2I, editing, and joint tasks. After high-quality SFT and reinforcement learning, DreamLite achieves GenEval (0.72) for image generation and ImgEdit (4.11) for image editing, outperforming existing on-device models and remaining competitive with several server-side models. By employing step distillation, we further reduce denoising processing to just 4 steps, enabling our DreamLite could generate or edit a 1024 x 1024 image in less than 1s on a Xiaomi 14 smartphone. To the best of our knowledge, DreamLite is the first unified on-device diffusion model that supports both image generation and image editing.
title DreamLite: A Lightweight On-Device Unified Model for Image Generation and Editing
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.28713