Generative AI art, which allows users to generate images based on descriptions, has become popular but can be slow. However, researchers at Tsinghua University and Hugging Face have developed a technique called LCM-LoRA that brings generative AI art creation into real-time.
Users can now paint simple drawings or shapes, and AI art creation applications will render new art instantaneously. The technique works for both 2D and 3D images, which could have applications in mixed reality, games, and film production. This advancement has been hailed as a game-changer by the AI arts community, and it is expected to lead to new user experiences. LCM-LoRA is a universal training-free acceleration module that speeds up the process of generating AI artwork, reducing the required sampling steps and enabling faster results in real-time.