# @nvidia on YouTube

- **Type:** Video
- **Original URL:** https://youtube.com/watch?v=1QPh70Es_oU
- **Gondola URL:** https://gondola.cc/posts/68415086-nvidia-youtube
- **Thumbnail:** https://img.gondola.cc/tr:w-,h-,fo-auto/postThumbnails/7c45413631.jpg
- **Posted:** 2026-07-20T15:01:49.000+00:00
- **Account Owner:** NVIDIA (@nvidia) — https://gondola.cc/nvidia

## Caption

The real world is infinite and unpredictable. Physical AI needs data—but real-world data is impossible to scale.

Meet NVIDIA Cosmos—the frontier of physical AI. Built on a new mixture-of-transformers architecture, this open omnimodel brings together pixels, actions, sound, and language to reason, plan, and generate what comes next. 

Developers can post-train Cosmos across embodiments and use cases. It can understand the physical world, generate physics-aware synthetic video, and simulate plausible futures and with post-training- Cosmos becomes a world action model that perceives, reasons, plans, and generates actions. Now available in multiple sizes.

📖 Read the Blog → https://nvda.ws/4vUMP5D
💻 Learn More About Cosmos → https://nvda.ws/4gDED5D
🤗 Try it on Hugging Face → https://huggingface.co/collections/nvidia/cosmos3

#SIGGRAPH2026 #NVIDIACosmos #PhysicalAI #Robotics #AutonomousVehicles #VisionAI

## Stats

- **Views:** 6,463
- **Likes:** 214
- **Shares:** 0
- **Comments:** 0

## Tags

autonomousvehicles, visionai, siggraph2026, physicalai, nvidiacosmos, robotics

---
Copyright (c) Gondola