Recent advancements in diffusion models are making significant strides across various applications. At SIGGRAPH Asia 2024, Gradio showcased the Instant Drag tool, enabling fast, photo-realistic edits in images while maintaining content integrity. Disney Research introduced the Robot Motion Diffusion Model (RobotMDM), a text-conditioned kinematic diffusion model for robotic motion generation. Princeton researchers unveiled RNAdiffusion, a latent diffusion model for generating and optimizing RNA sequences. Additionally, ViewCrafter can generate high-quality 3D views from single images using a video diffusion model. The OmniGen model was also introduced, offering unified image generation without the need for additional modules like ControlNet. Another notable development is DepthCrafter, which generates high-quality depth map sequences for videos using a three-stage training method. TurboEdit was introduced for simple text-based image editing in 1/2 sec using few-step diffusion models. RealVisXL V5.0 Lightning was highlighted in image outpainting using ControlNet and Stable Diffusion XL models.