Loading video...

Video Failed to Load

Go Home

Multiple modalities, many applications. Cosmos 3, the first omni-model for physical AI, is here. Check out some amazing examples below 👇

NVIDIAAI's profile picture

NVIDIA AI

325,604 subscribers

23,777 views • 2 months ago •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Introducing Cosmos 3: Our latest frontier model for Physical AI Cosmos 3 is the world’s first fully open omnimodel with native vision reasoning, world and action generation. Today we’re releasing Super (32B) and Nano (8B) variants.
3:13

Introducing Cosmos 3: Our latest frontier model for Physical AI Cosmos 3 is the world’s first fully open omnimodel with native vision reasoning, world and action generation. Today we’re releasing Super (32B) and Nano (8B) variants.

NVIDIA AI

425,707 views • 2 months ago

"AI Generated 3D is not production ready." The AI Generated 3D: 👇🙃 -- more (amazing) examples below. #GenAI #3D #GameArt
0:50

"AI Generated 3D is not production ready." The AI Generated 3D: 👇🙃 -- more (amazing) examples below. #GenAI #3D #GameArt

Emm | scenario.com

331,216 views • 2 years ago

The first generative AI for voice creation is here! We call it Voice Design - check out the quick demo below Try it out yourself: Read more on our blog:
0:48

The first generative AI for voice creation is here! We call it Voice Design - check out the quick demo below Try it out yourself: Read more on our blog:

ElevenLabs

84,658 views • 3 years ago

Breaking news: Cosmos 3 is here. They are attempting to do something completely new 🤯 Why is Physical AI much harder than building a chatbot? Understanding the world is not enough, robots need to predict it and act inside it. That's the idea behind NVIDIA Cosmos 3: → Reasoning model understands what's happening from video, images, text, and actions. → World model generates future states of the environment. → Action model generates the actions needed to achieve a goal. Previous systems often stitched these capabilities together using separate models. Cosmos 3 combines them into a single architecture with two components: ▪️ Reasoner Tower Analyzes observations and builds an understanding of objects, motion, interactions, and physical context. ▪️ Generator Tower Uses that understanding to generate future videos and action sequences that obey physical constraints. So Cosmos 3 moves from: Perception → Model A Prediction → Model B Actions → Model C to: Perception + Prediction + Actions → One unified system. The goal is to make robots, autonomous vehicles, and smart environments better at answering three questions: 1. What is happening? 2. What will happen next? 3. What should I do? That's a big shift from today's AI, which mostly focuses on generating text. And check the benchmarks! Physical AI needs to generate decisions that survive contact with the real world. 🚗🤖
0:33

Breaking news: Cosmos 3 is here. They are attempting to do something completely new 🤯 Why is Physical AI much harder than building a chatbot? Understanding the world is not enough, robots need to predict it and act inside it. That's the idea behind NVIDIA Cosmos 3: → Reasoning model understands what's happening from video, images, text, and actions. → World model generates future states of the environment. → Action model generates the actions needed to achieve a goal. Previous systems often stitched these capabilities together using separate models. Cosmos 3 combines them into a single architecture with two components: ▪️ Reasoner Tower Analyzes observations and builds an understanding of objects, motion, interactions, and physical context. ▪️ Generator Tower Uses that understanding to generate future videos and action sequences that obey physical constraints. So Cosmos 3 moves from: Perception → Model A Prediction → Model B Actions → Model C to: Perception + Prediction + Actions → One unified system. The goal is to make robots, autonomous vehicles, and smart environments better at answering three questions: 1. What is happening? 2. What will happen next? 3. What should I do? That's a big shift from today's AI, which mostly focuses on generating text. And check the benchmarks! Physical AI needs to generate decisions that survive contact with the real world. 🚗🤖

Turing Post

13,836 views • 2 months ago

RIP Hollywood! Moonvalley just dropped Marey, the world’s first AI video model built for professional production. Step into the director’s chair with precision control, clean data, and cinematic output. 10 examples below + link to try:👇
0:21

RIP Hollywood! Moonvalley just dropped Marey, the world’s first AI video model built for professional production. Step into the director’s chair with precision control, clean data, and cinematic output. 10 examples below + link to try:👇

Angry Tom

89,573 views • 1 year ago

We just released the world's first open-source humanoid physical strike AI! The AI is trained with large-scale self-play RL with a pre-trained ASE model from Jason Peng. Thank ZiyiBird for this amazing work.
1:11

We just released the world's first open-source humanoid physical strike AI! The AI is trained with large-scale self-play RL with a pre-trained ASE model from Jason Peng. Thank ZiyiBird for this amazing work.

Flood Sung

101,777 views • 3 years ago

🔥 We fine-tuned Gemma 3 using Unsloth AI on a custom instruction dataset scraped and built with . The model is now able to answer questions about OpenAI Agents SDK documentation. Check out the demo below👇
0:48

🔥 We fine-tuned Gemma 3 using Unsloth AI on a custom instruction dataset scraped and built with . The model is now able to answer questions about OpenAI Agents SDK documentation. Check out the demo below👇

Eric Ciarla (hiring)

28,794 views • 1 year ago

Our team is proud to announce that the first major update for #enshrouded is finally here!!! 🎉🎉🎉 Check out our highlights video below to get all the juicy details 👇
0:46

Our team is proud to announce that the first major update for #enshrouded is finally here!!! 🎉🎉🎉 Check out our highlights video below to get all the juicy details 👇

Enshrouded by Keen Games

112,220 views • 2 years ago

This is crazy... Skywork AI just dropped SkyReels-V2, the world's first open source AI video model that supports unlimited duration. Create infinite-length cinematic videos, straight from your browser. More examples below:
0:28

This is crazy... Skywork AI just dropped SkyReels-V2, the world's first open source AI video model that supports unlimited duration. Create infinite-length cinematic videos, straight from your browser. More examples below:

Angry Tom

302,751 views • 1 year ago

you're cooked if you're still doing single-character swapping and motion control. multiple character swaps + motion control - ONLY on Viggle. yes, we're the only AI video model that can do this. and we just launched our newest model V4 examples + tutorial + links below 👇
0:32

you're cooked if you're still doing single-character swapping and motion control. multiple character swaps + motion control - ONLY on Viggle. yes, we're the only AI video model that can do this. and we just launched our newest model V4 examples + tutorial + links below 👇

ViggleAI

59,699 views • 5 months ago

This AI animation workflow is making a splash 💦 GPT Image 2 is available as a partner AI model in Adobe Firefly and it's perfect for animations like this. Check out the workflow and full prompts below 👇 #AdobeFireflyAmbassadors #Ad #AdobePartnerModels
1:38

This AI animation workflow is making a splash 💦 GPT Image 2 is available as a partner AI model in Adobe Firefly and it's perfect for animations like this. Check out the workflow and full prompts below 👇 #AdobeFireflyAmbassadors #Ad #AdobePartnerModels

proper

68,385 views • 3 months ago

We build Cosmos-Predict2 as a world foundation model for Physical AI builders — fully open and adaptable. Post-train it for specialized tasks or different output types. Available in multiple sizes, resolutions, and frame rates. 📷 Watch the repo walkthrough ⚒️ Visit for more #NVIDIACosmos #PhysicalAI
3:26

We build Cosmos-Predict2 as a world foundation model for Physical AI builders — fully open and adaptable. Post-train it for specialized tasks or different output types. Available in multiple sizes, resolutions, and frame rates. 📷 Watch the repo walkthrough ⚒️ Visit for more #NVIDIACosmos #PhysicalAI

Hanzi Mao

32,000 views • 1 year ago

It’s finally here. Boba Anime 1.4 🎬 This new release takes anime generation to another level. Huge shoutout to Boba AI Labs for setting a new standard in AI anime creation. Watch the launch video below, then check out 6 examples of what Boba Anime 1.4 can do 👇
0:54

It’s finally here. Boba Anime 1.4 🎬 This new release takes anime generation to another level. Huge shoutout to Boba AI Labs for setting a new standard in AI anime creation. Watch the launch video below, then check out 6 examples of what Boba Anime 1.4 can do 👇

Kiran Bharambe

61,108 views • 10 months ago

Introducing Mothership, the first workspace for AI agents. Mothership is the central intelligence layer for your AI workforce. Autonomous agents, fully observable and editable. Check out what Mothership can do below.
2:15

Introducing Mothership, the first workspace for AI agents. Mothership is the central intelligence layer for your AI workforce. Autonomous agents, fully observable and editable. Check out what Mothership can do below.

Emir Karabeg

170,464 views • 4 months ago

AI TEXT-TO-VIDEO IS SERIOUSLY IMPRESSIVE In the heated battle for market dominance, Runway released its latest version, Gen-3. Here are some examples. Source: Ole Lehmann
0:10

AI TEXT-TO-VIDEO IS SERIOUSLY IMPRESSIVE In the heated battle for market dominance, Runway released its latest version, Gen-3. Here are some examples. Source: Ole Lehmann

Mario Nawfal

34,785 views • 2 years ago

Galactic Gladiator's Goredrake Mount in Midnight! #worldofwarcraft This is the Season 1 Gladiator mount! The model has multiple colors, including a purple one. Maybe it’s from one of the later seasons? Check out the preview below! 👇
0:26

Galactic Gladiator's Goredrake Mount in Midnight! #worldofwarcraft This is the Season 1 Gladiator mount! The model has multiple colors, including a purple one. Maybe it’s from one of the later seasons? Check out the preview below! 👇

Leyst

84,041 views • 8 months ago

This is THE moment of Physical AI! We are officially announcing Cosmos 3: Omnimodal World Models for Physical AI 🚀 - Cosmos 3 is an omnimodal world model: within a unified architecture, it can understand and generate language, images, video, audio, and actions. - It is not just a VLM, not just a video generator, not just an audio-visual generative model, and not just a physics simulator / world-action model. It can understand images and videos, generate images, videos, and audio, simulate future worlds, predict actions, and generate robot policies—enabling models to truly begin to “touch the world.” - Cosmos 3 is the #1 open-weight reasoner / T2I / I2V / robot policy across many benchmarks. Huge thanks to every teammate who fought side by side on this journey—from architecture, data, training, infra, serving, and evaluation to post-training. Every part of this project carries an incredible amount of hard work. This was my first time leading a project as Tech Lead, and I feel truly fortunate. The future of Physical AI needs models that can not only “see” and “describe” the world, but also “imagine,” “simulate,” and “act”—and eventually close the loop with the real world. I hope Cosmos 3 can become an important starting point for this direction, and I’m excited to push Physical AI into its next stage together with the open-source community. Welcome to the era of Physical AI. HuggingFace: Project Website: Code:
0:44

This is THE moment of Physical AI! We are officially announcing Cosmos 3: Omnimodal World Models for Physical AI 🚀 - Cosmos 3 is an omnimodal world model: within a unified architecture, it can understand and generate language, images, video, audio, and actions. - It is not just a VLM, not just a video generator, not just an audio-visual generative model, and not just a physics simulator / world-action model. It can understand images and videos, generate images, videos, and audio, simulate future worlds, predict actions, and generate robot policies—enabling models to truly begin to “touch the world.” - Cosmos 3 is the #1 open-weight reasoner / T2I / I2V / robot policy across many benchmarks. Huge thanks to every teammate who fought side by side on this journey—from architecture, data, training, infra, serving, and evaluation to post-training. Every part of this project carries an incredible amount of hard work. This was my first time leading a project as Tech Lead, and I feel truly fortunate. The future of Physical AI needs models that can not only “see” and “describe” the world, but also “imagine,” “simulate,” and “act”—and eventually close the loop with the real world. I hope Cosmos 3 can become an important starting point for this direction, and I’m excited to push Physical AI into its next stage together with the open-source community. Welcome to the era of Physical AI. HuggingFace: Project Website: Code:

Max Zhaoshuo Li 李赵硕

1,078,418 views • 2 months ago

EXPERIMENT: VFX + Camera Transitions in Veo 3 I've been playing with VEO 3, trying out some cool VFX and camera transitions with simple prompts. Check out these 6 examples (plus all the prompts) ... 👇
0:41

EXPERIMENT: VFX + Camera Transitions in Veo 3 I've been playing with VEO 3, trying out some cool VFX and camera transitions with simple prompts. Check out these 6 examples (plus all the prompts) ... 👇

Marco

66,980 views • 1 year ago

The Next Generation of Payments Is Here! Today, Razorpay becomes India’s first payment gateway with an Official MCP (Model Context Protocol) Server, designed for an AI-first world. Here is how it works 👇
0:37

The Next Generation of Payments Is Here! Today, Razorpay becomes India’s first payment gateway with an Official MCP (Model Context Protocol) Server, designed for an AI-first world. Here is how it works 👇

Razorpay

158,182 views • 1 year ago