A new distribution matching distillation (DMD) technique merges GAN principles with diffusion models, achieving 30x faster high-quality image generation in a single computational step and enhancing tools like Stable Diffusion and DALL-E.

MIT is a renowned institution for education and research, offering insights into science, engineering, and technology. Through publications, research papers, and academic programs, MIT's platform provides insights into  research, innovation, and education in various fields. Students, researchers, and technology enthusiasts can learn about MIT's contributions to science and technology and explore opportunities for academic and professional development.

MIT News

Researchers at MIT have introduced a new framework that allows for the generation of high-quality images 30 times faster than traditional diffusion models. The framework, called distribution matching distillation (DMD), simplifies the multi-step process into a single step and retains the quality of the generated images. This advancement has potential applications in design tools, drug discovery, and 3D modeling.

AI generates high-quality images 30 times faster in a single step