Some links may be affiliate links. We may earn a small commission at no extra cost to you. Learn more
NVIDIA Cosmos generates photorealistic virtual worlds from text/video for robot training using 20M hours pre-trained data.
Editor's take: “Spatial/world generation is an emerging and exciting capability” — Sohail Akhtar
Spatial/world generation is an emerging and exciting capability
Completely free for research and development with NVIDIA hardware optimization.
Associated Tags
nvidia cosmos ai, virtual worlds generation, robot training environments, 20m hours driving data, synthetic data generation, autonomous vehicle ai, omniverse integration, text to 3d world
Multimodal AI world model by World Labs that generates persistent, navigable 3D environments from text, images, video, or 3D layouts, with in-scene editing and Gaussian splat, mesh, and video export.
Freepik's ultra-realistic text-to-image generator features real-time editing, infinite canvas scrolling, and professional design tools.
Open-source AI model by Tencent that generates explorable, interactive 3D worlds from text or image inputs using panoramic scene reconstruction.
NVIDIA research model that generates textured, production-ready 3D assets with PBR materials from text or image inputs in around two minutes.
