NVIDIA Cosmos: Open-Source World Models for Physical AI (10K

What if you could predict how the physical world behaves — not by simulating physics equations, but by learning from the world itself?

NVIDIA Cosmos is exactly that: an open-source platform of world models trained to understand and generate the physical world. It doesn’t just generate images of a robot moving — it predicts the physics, the timing, the cause-and-effect of that motion.

Cosmos 3 is NVIDIA’s latest model family, built on a unified Mixture-of-Transformers (MoT) architecture that handles language, images, video, audio, and action sequences simultaneously. Two runtimes: a Reasoner (for world understanding and planning) and a Generator (for world simulation and synthetic data creation).

The models range from 16B (Nano) to 64B (Super) parameters, available on HuggingFace. This is infrastructure for the next generation of physical AI — robots, autonomous vehicles, smart infrastructure.

What Is NVIDIA Cosmos? #

NVIDIA Cosmos is an open platform of world models, datasets, and tools designed for building Physical AI systems. It goes beyond what traditional AI can do:

Traditional AI:         Cosmos:
  Input → Output      →  Input → Reasoning → Output
  (image in,          (understand physics,
   caption out)        predict future,
                       generate actions)

Key capabilities:

• World understanding: Analyze videos and images for captions, temporal events, next actions, spatial grounding, physical plausibility, and causal outcomes
• World generation: Produce images, videos, synchronized sound, and action-conditioned rollouts from text, image, video, or action inputs
• Action modeling: Predict policy actions, inverse dynamics, and forward dynamics for robotics, camera motion, egocentric motion, and autonomous driving

The Cosmos 3 model family includes:

| Model | Size | Capability | |

Model Family #

Getting Started #

# Pull a model from HuggingFace
git lfs install
git clone https://huggingface.co/nvidia/Cosmos-Nano-Reasoner

# Or use the NVIDIA API
import openai
client = openai.OpenAI(base_url="https://api.nvidia.com/v1")
resp = client.chat.completions.create(
    model="nvidia/cosmos-nano-reasoner",
    messages=[{"role": "user", "content": "Predict the next 5 seconds of this scene: [video]"}],
)

Use Cases #

• Robot training: generate synthetic training scenarios for embodied AI
• Autonomous vehicles: simulate rare edge cases safely
• Industrial inspection: predict equipment behavior before failures
• Embodied reasoning: give robots a “world model” for planning

Why World Models Matter #

Traditional AI learns patterns in data; world models learn causality in the physical world. A robot with a world model can predict the consequences of its actions — grasping, moving, avoiding — before executing them. Cosmos makes this capability open and accessible, which is why it is foundational infrastructure for physical AI in 2026.

Conclusion #

NVIDIA Cosmos is the most significant open-source step toward machines that understand physics, not just patterns. With the Cosmos 3 family (Nano to Super), the Reasoner/Generator split, and open weights on HuggingFace, it gives every robotics and simulation team access to world-model infrastructure that was previously locked inside research labs.