

NVIDIA Cosmos
#5 v 3D a generování světůnvidia · od Mai 31, 2026 · 7× · naposledy 30. 6. 2026
NVIDIA Cosmos 3 is a suite of omnimodal world foundation models designed to jointly process and generate language, images, video, audio, and action sequences within a unified Mixture-of-Transformers architecture. Cosmos is a world foundation model platform designed to accelerate the development of Physical AI by enabling machines to understand, simulate, and interact with the physical world across robotics, autonomous driving, and smart space environments. The model seamlessly unifies critical modalities for Physical AI—effectively subsuming vision-language models, video generators, world simulators, and world-action models into a single framework.
Vlastnosti
| Input Format | Text, image, video, audio, action sequences |
| Model Size | Edge (4B), Nano (16B), Super (64B) |
| Video Resolution | 256p, 480p, 720p |
| Video Length | 5–400 frames |