ComfyUI now supports Nvidia Cosmos

ComfyUI has rolled out support for Nvidia Cosmos, introducing groundbreaking tools for AI video generation. This update includes the Cosmos 7B model, optimized for high efficiency on GPUs with 24GB VRAM and even functional on 12GB GPUs, albeit with slower render times. By lowering hardware demands, the integration promises broader access to advanced AI video…

Read More

Gaussian Splatting Plugin for Nuke

Irrealix, the team behind the popular Gaussian Splatting (3DGS) plugins for After Effects and DaVinci Resolve, has announced another breakthrough: Gaussian Splatting for Nuke.  This groundbreaking plugin brings Radiance Field technology into Nuke’s industry-standard compositing environment, VFX professionals to import, manipulate, and render 3D Gaussian Splatting models in real-time with GPU acceleration. So What is Gaussian…

Read More

Custom nodes are the Heart and soul of ComfyUI

Let’s be real: custom nodes are the heart and soul of ComfyUI. There’s basically a node for everything—new open-source models, workflow sharing, gorgeous UI features—you name it. Heck, even ComfyUI-Manager is a custom node. But, with great customization comes great headaches. Custom nodes can cause messy dependency conflicts, eat up your storage with massive downloads,…

Read More

How AI Is Redefining VFX Careers

AI might reduce the number of people needed for traditional VFX tasks, but it’s also unlocking new creative horizons. One such role is the “FaceSwap Artist“, as recently advertised by Industrial Light & Magic (ILM). This job offers a glimpse into how AI is reshaping the landscape of VFX work. What Is a FaceSwap Artist?…

Read More

Meet Meshtron: Revolutionizing 3D Mesh Generation

Creating detailed 3D meshes just got a whole lot easier with Meshtron by Nvidia Research, a cutting-edge tool that combines smart architecture and innovative techniques to produce artist-quality 3D assets. What Makes Meshtron Special? Meshtron uses a design called the Hourglass Transformer, coupled with sliding window attention, to efficiently generate 3D meshes. Its clever approach…

Read More

Leaked Footage from Openai’s Sora 2

OpenAI is gearing up to revolutionize video creation with Sora, an AI model capable of generating 1-minute videos from text, text+image, or text+video prompts. The announcement was made by Chad Nelson during the C21Media Keynote in London, where he revealed that the technology would be available “very, very soon.” Sora’s capabilities, as showcased, are groundbreaking….

Read More