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

ArXi:2603.28713v1 Announce Type: new 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 for image editing. In this paper, we propose DreamLite, a compact unified on-device diffusion model (0.39B) that s both T2I generation and text-guided image editing within a single network.