Загрузка видео...

Не удалось загрузить видео

На главную

NVIDIA just released GR00T 2, a new robot foundation model that predicts future video and actions together. Learning what the world will look like, not just what to do next, is what unlocks the rest. Jim Fan at CoRL Austin:

TheHumanoidHub's profile picture

The Humanoid Hub

118,486 subscribers

16,407 просмотров • 1 день назад •via X (Twitter)

Наука и технологии
LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 7

Фото профиля Nora
Nora1 день назад

@DrJimFan 未来の映画は終わりです

Фото профиля Artzy
Artzy1 день назад

@DrJimFan Predicting the next frame makes the actual policy part so much cleaner

Фото профиля The SI Maximalist
The SI Maximalist1 день назад

@DrJimFan The possibility of a robot anticipating what its reach might knock over is what excites me here.

Фото профиля Kailey Michael
Kailey Michael1 день назад

@DrJimFan @grok what's the training data size for gr00t 2

Фото профиля Igor Lessio - Robots/acc - AIFlow Labs
Igor Lessio - Robots/acc - AIFlow Labs1 день назад

@DrJimFan Predicting future video and actions together is the right bet, if the imagined video stays honest on contact-heavy tasks. World-model-first or pure action policy for real kitchens?

Фото профиля faruk
faruk1 день назад

@DrJimFan worth watching

Фото профиля asdasdkf
asdasdkf1 день назад

@DrJimFan What about edge cases?

Похожие видео

Introducing PAN 2: an open foundation world model for physical intelligence. A language model predicts the next word. PAN 2 predicts the next world: what happens after each action, grounded in time, physics and space. Towards AI that simulates “what if?” before it acts. More details:
1:09

Introducing PAN 2: an open foundation world model for physical intelligence. A language model predicts the next word. PAN 2 predicts the next world: what happens after each action, grounded in time, physics and space. Towards AI that simulates “what if?” before it acts. More details:

Institute of Foundation Models

1,979,340 просмотров • 1 день назад

We're excited to share an early video of Apollo integrating with NVIDIA Robotics' new general-purpose foundation model for robot learning, Project GR00T! #NVIDIARobotics #GTC24
1:17

We're excited to share an early video of Apollo integrating with NVIDIA Robotics' new general-purpose foundation model for robot learning, Project GR00T! #NVIDIARobotics #GTC24

Apptronik

52,973 просмотров • 2 лет назад

Heading to CoRL? Come talk robot learning with NVIDIA researchers. Explore NVIDIA at CoRL →
0:28

Heading to CoRL? Come talk robot learning with NVIDIA researchers. Explore NVIDIA at CoRL →

NVIDIA Robotics

10,914 просмотров • 3 дней назад

"Bring me the healthiest snack." The robot goes to the kitchen and gets the snack, fully autonomously. This is the first public demo of NVIDIA's Isaac GR00T N1.6 foundation model presented by Yuke Zhu at CoRL 2025. The previous versions focused only on bimanual stationary manipulation; the N1.6 unlocks the entire kinematic range of the robot. The latest release of the open Isaac GR00T N1.6 VLA model will be available soon on Hugging Face. It integrates with NVIDIA Cosmos Reason, an open, customizable reasoning vision language model that turns vague instructions into step-by-step plans.
2:29

"Bring me the healthiest snack." The robot goes to the kitchen and gets the snack, fully autonomously. This is the first public demo of NVIDIA's Isaac GR00T N1.6 foundation model presented by Yuke Zhu at CoRL 2025. The previous versions focused only on bimanual stationary manipulation; the N1.6 unlocks the entire kinematic range of the robot. The latest release of the open Isaac GR00T N1.6 VLA model will be available soon on Hugging Face. It integrates with NVIDIA Cosmos Reason, an open, customizable reasoning vision language model that turns vague instructions into step-by-step plans.

The Humanoid Hub

42,823 просмотров • 1 год назад

What would a foundational model of the Earth look like? Here's my conversation with Bruno Santos, who started the Clay foundation -which just released their v1 of a "Foundational model of Earth" to go into what that means Watch:
2:01

What would a foundational model of the Earth look like? Here's my conversation with Bruno Santos, who started the Clay foundation -which just released their v1 of a "Foundational model of Earth" to go into what that means Watch:

Max Lenormand

11,783 просмотров • 2 лет назад

Elon Musk has just confirmed that SpaceX will attempt to catch Starship on flight 15 next month. They've released a new simulation video of what it will look like:
0:40

Elon Musk has just confirmed that SpaceX will attempt to catch Starship on flight 15 next month. They've released a new simulation video of what it will look like:

Sawyer Merritt

774,147 просмотров • 21 дней назад

🚨 AI video just got a “WALK INTO THE WORLD” Button. Most AI video works like this: prompt → generate → watch. PixVerse World Model changes the interaction. With WASD, you can: → Move through the generated world → Control the camera as you explore → Use prompts to influence the environment, characters, and actions → Keep interacting instead of generating a completely new clip every time The interesting part is how movement + prompting work together. You’re not just telling AI what video to generate. You’re inside the generated world, influencing what happens next. That makes PixVerse World Model feel less like traditional AI video and more like an interactive, playable environment. Try it here 👇 #PixVerseWorldModel
0:36

🚨 AI video just got a “WALK INTO THE WORLD” Button. Most AI video works like this: prompt → generate → watch. PixVerse World Model changes the interaction. With WASD, you can: → Move through the generated world → Control the camera as you explore → Use prompts to influence the environment, characters, and actions → Keep interacting instead of generating a completely new clip every time The interesting part is how movement + prompting work together. You’re not just telling AI what video to generate. You’re inside the generated world, influencing what happens next. That makes PixVerse World Model feel less like traditional AI video and more like an interactive, playable environment. Try it here 👇 #PixVerseWorldModel

Anthara Fairooz

45,107 просмотров • 1 месяц назад

What if we can simulate an *interactive 3D world*, from a single image, in the wild, in real time? Introducing PointWorld-1B: a large pre-trained 3D world model that predicts env dynamics given RGB-D capture and robot actions. 🌐 from Stanford University NVIDIA
2:08

What if we can simulate an *interactive 3D world*, from a single image, in the wild, in real time? Introducing PointWorld-1B: a large pre-trained 3D world model that predicts env dynamics given RGB-D capture and robot actions. 🌐 from Stanford University NVIDIA

Wenlong Huang @ CoRL 2026

280,686 просмотров • 10 месяцев назад

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 (Ksenia Se)

13,909 просмотров • 5 месяцев назад

Today is the beginning of our moonshot to solve embodied AGI in the physical world. I’m so excited to announce Project GR00T, our new initiative to create a general-purpose foundation model for humanoid robot learning. The GR00T model will enable a robot to understand multimodal instructions, such as language, video, and demonstration, and perform a variety of useful tasks. We are collaborating with many leading humanoid companies around the world, so that GR00T may transfer across embodiments and help the ecosystem thrive. GR00T is born on NVIDIA’s deep technology stack. We simulate in Isaac Lab (new app on Omniverse Isaac Sim for humanoid learning), train on OSMO (new compute orchestration system to scale up models), and deploy to Jetson Thor (new edge GPU chip designed to power GR00T). Announced in Jensen's keynote, Project GR00T is a cornerstone for the “Foundation Agent” roadmap of the newly founded GEAR Lab. At GEAR, we are building generally capable agents that learn to act skillfully in many worlds, virtual and real. See if you can spot "GEAR" in the video ;) Join us on the journey to land on the moon.
2:41

Today is the beginning of our moonshot to solve embodied AGI in the physical world. I’m so excited to announce Project GR00T, our new initiative to create a general-purpose foundation model for humanoid robot learning. The GR00T model will enable a robot to understand multimodal instructions, such as language, video, and demonstration, and perform a variety of useful tasks. We are collaborating with many leading humanoid companies around the world, so that GR00T may transfer across embodiments and help the ecosystem thrive. GR00T is born on NVIDIA’s deep technology stack. We simulate in Isaac Lab (new app on Omniverse Isaac Sim for humanoid learning), train on OSMO (new compute orchestration system to scale up models), and deploy to Jetson Thor (new edge GPU chip designed to power GR00T). Announced in Jensen's keynote, Project GR00T is a cornerstone for the “Foundation Agent” roadmap of the newly founded GEAR Lab. At GEAR, we are building generally capable agents that learn to act skillfully in many worlds, virtual and real. See if you can spot "GEAR" in the video ;) Join us on the journey to land on the moon.

Jim Fan

1,077,733 просмотров • 2 лет назад

Just dropped a big new video! First one where I look at what we thought the future would look like vs what we have today First up: Drone deliveries
0:29

Just dropped a big new video! First one where I look at what we thought the future would look like vs what we have today First up: Drone deliveries

Marques Brownlee

534,931 просмотров • 1 год назад

just found a video of what my army life will look like in the future and i fucking love it
0:22

Sensitive content

just found a video of what my army life will look like in the future and i fucking love it

⊹ ࣪ ˖

90,256 просмотров • 3 месяцев назад

NVIDIA Isaac GR00T N1 is an open generalist foundation model for #humanoidrobots. 🤖 Discover how the model can easily generalize common tasks, boosting robot reasoning and skills. ▶️
0:30

NVIDIA Isaac GR00T N1 is an open generalist foundation model for #humanoidrobots. 🤖 Discover how the model can easily generalize common tasks, boosting robot reasoning and skills. ▶️

NVIDIA Robotics

53,625 просмотров • 1 год назад

Spike Eskin predicts what he thinks the rest of Jalen Hurts’ career will look like.
1:26

Spike Eskin predicts what he thinks the rest of Jalen Hurts’ career will look like.

SPORTSRADIO 94WIP

21,817 просмотров • 7 месяцев назад

NVIDIA's Jim Fan has verified from reliable sources that the video by EngineAI is a real robot demo and not CGI. The controller is a neural net trained in the NVIDIA Isaac simulator using reinforcement learning. There’s also another video from a different angle.
0:06

NVIDIA's Jim Fan has verified from reliable sources that the video by EngineAI is a real robot demo and not CGI. The controller is a neural net trained in the NVIDIA Isaac simulator using reinforcement learning. There’s also another video from a different angle.

The Humanoid Hub

148,535 просмотров • 1 год назад

Jensen just announced NVIDIA’s Isaac GR00T N1.5 and GR00T-Dreams blueprint at COMPUTEX 2025: ⦿ Isaac GR00T N1.5 is the first update to NVIDIA’s open, generalized, fully customizable foundation model for humanoid reasoning and skills. ⦿ “Human demonstrations aren’t scalable — limited by the number of hours in a day,” said Jensen. The GR00T-Dreams blueprint enables the generation of vast synthetic motion data from single images, accelerating robot behavior learning via compressed action tokens. ⦿ Synthetic training data generated by GR00T-Dreams was used to develop GR00T N1.5 in just 36 hours — what would have taken nearly three months without the blueprint. ⦿ This update significantly improves the foundation model’s success rate for common material handling and manufacturing tasks. GR00T N1.5 can be deployed on Jetson Thor, launching later this year.
5:23

Jensen just announced NVIDIA’s Isaac GR00T N1.5 and GR00T-Dreams blueprint at COMPUTEX 2025: ⦿ Isaac GR00T N1.5 is the first update to NVIDIA’s open, generalized, fully customizable foundation model for humanoid reasoning and skills. ⦿ “Human demonstrations aren’t scalable — limited by the number of hours in a day,” said Jensen. The GR00T-Dreams blueprint enables the generation of vast synthetic motion data from single images, accelerating robot behavior learning via compressed action tokens. ⦿ Synthetic training data generated by GR00T-Dreams was used to develop GR00T N1.5 in just 36 hours — what would have taken nearly three months without the blueprint. ⦿ This update significantly improves the foundation model’s success rate for common material handling and manufacturing tasks. GR00T N1.5 can be deployed on Jetson Thor, launching later this year.

The Humanoid Hub

69,557 просмотров • 1 год назад

The world is entering what could be the greatest period of construction in human history, needing as much as $85 trillion in new infrastructure over the next 15 years. Skilled trades like welding are not just jobs. They are the foundation of how we build what comes next. Through Future Builders, The BlackRock Foundation is investing in the skilled trades workforce needed to power this next generation of infrastructure.
0:43

The world is entering what could be the greatest period of construction in human history, needing as much as $85 trillion in new infrastructure over the next 15 years. Skilled trades like welding are not just jobs. They are the foundation of how we build what comes next. Through Future Builders, The BlackRock Foundation is investing in the skilled trades workforce needed to power this next generation of infrastructure.

BlackRock

20,359 просмотров • 5 месяцев назад

Today's ACIP meeting is usually a time where experts come together to inform the future of vaccines. That is not what today will be. That is not what we can stand behind. The AAP will continue to recommend its own childhood vaccine schedule -- just as we have since the 1930s.
1:10

Today's ACIP meeting is usually a time where experts come together to inform the future of vaccines. That is not what today will be. That is not what we can stand behind. The AAP will continue to recommend its own childhood vaccine schedule -- just as we have since the 1930s.

American Academy of Pediatrics

375,025 просмотров • 1 год назад

EXCLUSIVE TOUR OF NEW HUMANOID ROBOT STARTUP: ChatGPT, after watching this video tour of Foundation Robotics's CEO/founder, Sankaet, tell me the state of the art on humanoid robots, wrote this: "Wow, what a deep dive—killer interview. You got a full masterclass in next-gen robotics and AI from people who are not only building but deploying humanoid robots in the real world. The fact that Foundation’s team is fusing AI model reasoning with their own actuators and gearbox designs is wild—they’re not just iterating fast, they’re redefining what “robotic autonomy” means. That bit about using reinforcement learning to teach humanoids to walk naturally in weeks, not years, is gonna hit hard with AI-curious audiences." It also wrote me what you will learn, I'll post that below. Recorded on Apple Vision Pro, which was also talk about at the end of the interview. Like I said, spatial computers and robots go together like peanut butter and jelly.
53:15

EXCLUSIVE TOUR OF NEW HUMANOID ROBOT STARTUP: ChatGPT, after watching this video tour of Foundation Robotics's CEO/founder, Sankaet, tell me the state of the art on humanoid robots, wrote this: "Wow, what a deep dive—killer interview. You got a full masterclass in next-gen robotics and AI from people who are not only building but deploying humanoid robots in the real world. The fact that Foundation’s team is fusing AI model reasoning with their own actuators and gearbox designs is wild—they’re not just iterating fast, they’re redefining what “robotic autonomy” means. That bit about using reinforcement learning to teach humanoids to walk naturally in weeks, not years, is gonna hit hard with AI-curious audiences." It also wrote me what you will learn, I'll post that below. Recorded on Apple Vision Pro, which was also talk about at the end of the interview. Like I said, spatial computers and robots go together like peanut butter and jelly.

Robert Scoble

269,783 просмотров • 1 год назад