We propose DreamLite, a compact unified on-device diffusion model (0.39B) that supports both text-to-image 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. By employing step distillation, DreamLite achieves 4-step inference, generating or editing a 1024×1024 image in less than 5 seconds on an iPhone 17 Pro — fully on-device, no cloud required.
Overall architecture of DreamLite.
2026.03: 🎉🎉🎉 DreamLite is released! See our project page and paper.
Real-time generation & editing on iPhone 17 Pro — no cloud, fully on-device.
| Image Editing | Style Transfer | Background Change |
|---|---|---|
Note: If the videos do not render on GitHub, please visit our project page to view the full demos.
# Clone the repository
git clone https://github.com/carlofkl/DreamLite.git
cd DreamLite| Model | Params | Resolution | Steps | HuggingFace |
|---|---|---|---|---|
| DreamLite (Base) | 0.39B | 1024×1024 | 28 | 🤗 To be released |
| DreamLite (Distilled) | 0.39B | 1024×1024 | 4 | 🤗 To be released |
Quantitative comparison with state-of-the-art methods on generation and editing benchmarks.
Text-to-Image generation comparison.
Text-guided image editing comparison.
| Method | Params | GenEval ↑ | DPG ↑ | ImgEdit ↑ | GEdit-EN-Q ↑ |
|---|---|---|---|---|---|
| FLUX.1-Dev / Kontext | 12B | 0.67 | 84.0 | 3.76 | 6.79 |
| BAGEL | 7B | 0.82 | 85.1 | 3.42 | 7.20 |
| OmniGen2 | 4B | 0.80 | 83.6 | 3.44 | 6.79 |
| LongCat-Image / Edit | 6B | 0.87 | 86.6 | 4.49 | 7.55 |
| DeepGen1.0 | 2B | 0.83 | 84.6 | 4.03 | 7.54 |
| SANA-1.6B | 1.6B | 0.67 | 84.8 | - | - |
| SANA-0.6B | 0.6B | 0.64 | 83.6 | - | - |
| SnapGen++ (small) | 0.4B | 0.66 | 85.2 | - | - |
| VIBE | 1.6B | - | - | 3.85 | 7.28 |
| EditMGT | 0.96B | - | - | 2.89 | 6.33 |
| DreamLite (Ours) | 0.39B | 0.72 | 85.8 | 4.11 | 6.88 |
- Release paper on arXiv
- Release inference code
- Release model weights on HuggingFace
- Release online demo
- Release Android & iOS App
We thank the great work from SDXL, SnapGen, Qwen and TAESDXL. The work is under supervision from Prof. Wangmeng Zuo.
If our work assists your research, feel free to give us a star ⭐ or cite us using:
@article{feng2026dreamlite,
title={DreamLite: A Lightweight On-Device Unified Model for Image Generation and Editing},
author={Kailai Feng and Yuxiang Wei and Bo Chen and Yang Pan and Hu Ye and Songwei Liu and Chenqian Yan and Yuan Gao},
journal={arXiv preprint arXiv:2603.28713},
year={2026}
}