Cosmos3-Nano

video

Cosmos3 is a world foundation model that unifies understanding and generation within a single Mixture-of-Transformer (MoT) architecture. Two tightly coupled towers—a Reasoner (vision-language model) and a Generator (world simulator)—share latent representations so that structured perception directly grounds realistic, temporally consistent simulation.

Capabilities

No supported capabilities recorded.

Details

Model IDnvidia/Cosmos3-Nano
Provider DeepInfra
Creator NVIDIA
Family—
License—
Statusactive
Typevideo
Input modalitiestext
Output modalitiesvideo
Open weightYes
Parameters—
Architecture—
Max duration—
Max resolution—
Resolutions—
FPS—
Release date—
Sourceofficial

Endpoints

videoPOST
https://api.deepinfra.com/v1/openai/video

Pricing

$0.0350 / second (720p)

API

GET/v1/models/deepinfra/nvidia/Cosmos3-Nano