Загрузка видео...

Не удалось загрузить видео

На главную

JUST IN: Odyssey unveils Odyssey-3, an AI world model that can learn from visual data and apply its knowledge to robots, cars, drones and video games.

WhaleInsider's profile picture

Whale Insider

675,857 subscribers

41,140 просмотров • 1 день назад •via X (Twitter)

Новости и политикаОбразованиеНаука и технологии
LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 15

Фото профиля bigfud.sol 🌱
bigfud.sol 🌱1 день назад

big runner will be $LMEOW fulfilling GCR's prophecy 476MWyHnzcUc2jAnBxdqVBFZpArT3VGGwPsUHjnjA3GC

Фото профиля Ebrahim
Ebrahim1 день назад

duck simulator looks harder than my whole stack

Фото профиля 8lends
8lends1 день назад

AI playing gta damn

Фото профиля Scamcoin
Scamcoin1 день назад

finally a duck with utility

Фото профиля Md Fahim
Md Fahim1 день назад

AI that understands the physical world could be a massive leap for robotics and autonomy. 👀

Фото профиля Zen🎒
Zen🎒1 день назад

Huhhh

Фото профиля Harsh Rana
Harsh Rana23 часов назад

soon

Фото профиля Whale Fun
Whale Fun1 день назад

AI is getting scarier by the day

Фото профиля Genius💡💹🧲 🤖
Genius💡💹🧲 🤖1 день назад

The multimodal learning angle here could be huge for robotics and autonomous systems

Фото профиля AGTP
AGTP1 день назад

This foundation model approach is a huge leap for robotics and autonomous systems. We actually went deeper on this here:

Фото профиля BTC Whale
BTC Whale1 день назад

This is actually mind blowing stuff

Фото профиля Roberto BTC
Roberto BTC1 день назад

That’s so nice innovation

Фото профиля OKN
OKN1 день назад

Amazing

Фото профиля OGBTC
OGBTC1 день назад

Wow

Фото профиля Altered State
Altered State1 день назад

That’s crazy thanks for sharing

Похожие видео

Odyssey is now previewing Odyssey-3, a world model that it calls a Physics Agent! The model was designed to learn how the world behaves from visual observation and then apply that understanding to control both real and simulated systems. Odyssey-3 can run across different machines and games, with a small amount of task-specific experience teaching it each set of controls. The preview covers humanoids, robot arms, a car and a drone being controlled by an AI model, playable video games, and training runs for other AI agents.
0:30

Odyssey is now previewing Odyssey-3, a world model that it calls a Physics Agent! The model was designed to learn how the world behaves from visual observation and then apply that understanding to control both real and simulated systems. Odyssey-3 can run across different machines and games, with a small amount of task-specific experience teaching it each set of controls. The preview covers humanoids, robot arms, a car and a drone being controlled by an AI model, playable video games, and training runs for other AI agents.

🚨 AI News | TestingCatalog

23,779 просмотров • 1 день назад

wild that one world model can go from controlling robots to driving cars to playing video games 🤯
0:30

wild that one world model can go from controlling robots to driving cars to playing video games 🤯

Tiffany Fong

25,806 просмотров • 1 день назад

JUST IN: One model powering humanoids, robot arms, cars, drones and games is a fascinating direction. Odyssey-3 shows how broad pretraining can provide a shared foundation, with just a few hours of task-specific experience needed to adapt
0:48

JUST IN: One model powering humanoids, robot arms, cars, drones and games is a fascinating direction. Odyssey-3 shows how broad pretraining can provide a shared foundation, with just a few hours of task-specific experience needed to adapt

Insider Paper

16,140 просмотров • 1 день назад

Today we’re unveiling Odyssey-3, a big step forward for foundation world models. It can control robots, power humanoids, drive cars (on the roads of India!), train AIs, pilot drones, and even play video games. We can’t wait to see what intelligent systems it enables.
1:36

Today we’re unveiling Odyssey-3, a big step forward for foundation world models. It can control robots, power humanoids, drive cars (on the roads of India!), train AIs, pilot drones, and even play video games. We can’t wait to see what intelligent systems it enables.

Odyssey

2,107,394 просмотров • 1 день назад

I'm excited to share Odyssey, my new thing! 🍿 We're building Hollywood-grade visual AI, where beautiful scenery, characters, lighting, and motion can be both generated and directed. Our mission is to deliver a better way to create movies, TV shows, and video games.
1:17

I'm excited to share Odyssey, my new thing! 🍿 We're building Hollywood-grade visual AI, where beautiful scenery, characters, lighting, and motion can be both generated and directed. Our mission is to deliver a better way to create movies, TV shows, and video games.

Oliver Cameron

221,994 просмотров • 2 лет назад

Featherweight Games is proud to announce Botworld Odyssey, an open world creature collecting RPG, coming to Nintendo Switch and Steam on October 24th, 2024. We are really excited about this and hope you can join us on this odyssey. Thank you for your support! #BotworldOdyssey
1:36

Featherweight Games is proud to announce Botworld Odyssey, an open world creature collecting RPG, coming to Nintendo Switch and Steam on October 24th, 2024. We are really excited about this and hope you can join us on this odyssey. Thank you for your support! #BotworldOdyssey

Botworld Adventure®

12,483 просмотров • 2 лет назад

With enough data, robots and AI can learn “world models” that let them predict the results of their actions. These models are a way to learn how embodied AI agents can perform a wide variety of useful tasks — but they require a huge amount of data. The team at General Intuition General Intuition has a solution: use data from video games! Games teach movement, problem solving, and complex spatial reasoning, and they come in a staggering diversity of forms, covering a wide variety of problems. What’s more, the captured data is high-quality, without the noise or annotation error that can come from We sat down with Pim de Witte and Adam Jelley from the General Intuition team to learn more about their history, their plans, and their philosophy.
53:51

With enough data, robots and AI can learn “world models” that let them predict the results of their actions. These models are a way to learn how embodied AI agents can perform a wide variety of useful tasks — but they require a huge amount of data. The team at General Intuition General Intuition has a solution: use data from video games! Games teach movement, problem solving, and complex spatial reasoning, and they come in a staggering diversity of forms, covering a wide variety of problems. What’s more, the captured data is high-quality, without the noise or annotation error that can come from We sat down with Pim de Witte and Adam Jelley from the General Intuition team to learn more about their history, their plans, and their philosophy.

RoboPapers

85,927 просмотров • 10 месяцев назад

Gemini 3 Pro Image (Nano Banana Pro) is available on Vertex AI and coming soon to Gemini Enterprise! Nano Banana Pro excels in visual design, world knowledge, and text generation. Learn more →
0:32

Gemini 3 Pro Image (Nano Banana Pro) is available on Vertex AI and coming soon to Gemini Enterprise! Nano Banana Pro excels in visual design, world knowledge, and text generation. Learn more →

Google Cloud

28,092 просмотров • 10 месяцев назад

Assassin's Creed Odyssey has a stunning and peaceful environment. Sometimes I just stop and enjoy the beautiful scenery. Ubisoft always nails world building in their games.
0:42

Assassin's Creed Odyssey has a stunning and peaceful environment. Sometimes I just stop and enjoy the beautiful scenery. Ubisoft always nails world building in their games.

Zephryss⛩️

38,406 просмотров • 6 месяцев назад

Our vision is for AI that uses world models to adapt in new and dynamic environments and efficiently learn new skills. We’re sharing V-JEPA 2, a new world model with state-of-the-art performance in visual understanding and prediction. V-JEPA 2 is a 1.2 billion-parameter model, trained on video, that can enable zero-shot planning in robots—allowing them to plan and execute tasks in unfamiliar environments. Learn more about V-JEPA 2 ➡️ As we continue working toward our goal of achieving advanced machine intelligence (AMI), we’re also releasing three new benchmarks for evaluating how well existing models can reason about the physical world from video. Learn more and download the new benchmarks ➡️
2:03

Our vision is for AI that uses world models to adapt in new and dynamic environments and efficiently learn new skills. We’re sharing V-JEPA 2, a new world model with state-of-the-art performance in visual understanding and prediction. V-JEPA 2 is a 1.2 billion-parameter model, trained on video, that can enable zero-shot planning in robots—allowing them to plan and execute tasks in unfamiliar environments. Learn more about V-JEPA 2 ➡️ As we continue working toward our goal of achieving advanced machine intelligence (AMI), we’re also releasing three new benchmarks for evaluating how well existing models can reason about the physical world from video. Learn more and download the new benchmarks ➡️

AI at Meta

310,561 просмотров • 1 год назад

Robots are getting smarter, but most still fail the same way. They don’t learn from their own mistakes. A new paper proposes something different: a way for robots to self-improve directly from their failures in the real world. It’s called PLD (Probe, Learn, Distill). The idea: instead of collecting endless human demos, let the robot figure out where it fails, learn how to recover, and then distill that knowledge back into its main model. Key takeaways from the research: ✅ Uses residual reinforcement learning to recover from policy failures ✅ Achieves 99% success on LIBERO and 100% on real Franka and YAM arms ✅ Runs hour-long manipulation tasks without human resets ✅ Builds a feedback loop between real-world data and model adaptation Unlike supervised fine-tuning, which relies on humans, PLD learns from the robot’s own experience. By training on its own failure distribution, the model becomes both more efficient and more aligned with the real world. It’s not just a technical shift, it’s a step toward robots that improve themselves through real-world practice. Thanks for sharing, Wenli Xiao !! Paper and demos: —- Weekly robotics and AI insights. Subscribe free:
0:47

Robots are getting smarter, but most still fail the same way. They don’t learn from their own mistakes. A new paper proposes something different: a way for robots to self-improve directly from their failures in the real world. It’s called PLD (Probe, Learn, Distill). The idea: instead of collecting endless human demos, let the robot figure out where it fails, learn how to recover, and then distill that knowledge back into its main model. Key takeaways from the research: ✅ Uses residual reinforcement learning to recover from policy failures ✅ Achieves 99% success on LIBERO and 100% on real Franka and YAM arms ✅ Runs hour-long manipulation tasks without human resets ✅ Builds a feedback loop between real-world data and model adaptation Unlike supervised fine-tuning, which relies on humans, PLD learns from the robot’s own experience. By training on its own failure distribution, the model becomes both more efficient and more aligned with the real world. It’s not just a technical shift, it’s a step toward robots that improve themselves through real-world practice. Thanks for sharing, Wenli Xiao !! Paper and demos: —- Weekly robotics and AI insights. Subscribe free:

Ilir Aliu

22,881 просмотров • 10 месяцев назад

Today, we announced 𝗥𝗧-𝟮: a first of its kind vision-language-action model to control robots. 🤖 It learns from both web and robotics data and translates this knowledge into generalised instructions. Find out more:
0:26

Today, we announced 𝗥𝗧-𝟮: a first of its kind vision-language-action model to control robots. 🤖 It learns from both web and robotics data and translates this knowledge into generalised instructions. Find out more:

Google DeepMind

537,902 просмотров • 3 лет назад

🤔What does it take to train an omni world model from scratch? 📽️Petabytes of video. 6⃣pipeline stages. And every improvement in data quality pays off twice when you're training a model that both generates and understands video. Our Reka Labs' data team on how it works → #PhysicalAI #WorldModels #AI #RekaLabs
0:38

🤔What does it take to train an omni world model from scratch? 📽️Petabytes of video. 6⃣pipeline stages. And every improvement in data quality pays off twice when you're training a model that both generates and understands video. Our Reka Labs' data team on how it works → #PhysicalAI #WorldModels #AI #RekaLabs

Reka

15,242 просмотров • 2 месяцев назад

With Diambra, you can create - and own - an AI Agent that learns to master the toughest video games 🎮 Learn more: 🔴 ⚫️
0:28

With Diambra, you can create - and own - an AI Agent that learns to master the toughest video games 🎮 Learn more: 🔴 ⚫️

Diambra

39,369 просмотров • 1 год назад

Robots can bake bread, cut carrots, and do karate kicks in demo videos. But how capable are they, really? Robocurve PBC measures how good AI and robots are in the physical world, from making a sandwich to building data centers.
1:37

Robots can bake bread, cut carrots, and do karate kicks in demo videos. But how capable are they, really? Robocurve PBC measures how good AI and robots are in the physical world, from making a sandwich to building data centers.

Jay Chooi

42,739 просмотров • 1 месяц назад

Robots might learn better from video than from language! 📼 Most Vision-Language-Action (VLA) models learn what to do from text, but still struggle with how things move in the real world. That makes them data-hungry and slow to train. mimic video takes a different route. Instead of grounding robot control in text, it grounds it in video, using large pre-trained video models that already capture physical motion and dynamics. The idea is straightforward: let the video model handle “what will happen next,” and let a smaller control model focus only on turning that visual plan into robot actions. The result is big gains in practice. Robots trained this way need 10× less data, converge twice as fast, and perform better on both simulated benchmarks and real bimanual manipulation tasks. If robots can “imagine” motion using video, control becomes a much simpler problem. Shoutout to Jonas Pai, Liam Achenbach, Oier Mees, Elvis Nava and the rest of the team! Here's the project page: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
0:47

Robots might learn better from video than from language! 📼 Most Vision-Language-Action (VLA) models learn what to do from text, but still struggle with how things move in the real world. That makes them data-hungry and slow to train. mimic video takes a different route. Instead of grounding robot control in text, it grounds it in video, using large pre-trained video models that already capture physical motion and dynamics. The idea is straightforward: let the video model handle “what will happen next,” and let a smaller control model focus only on turning that visual plan into robot actions. The result is big gains in practice. Robots trained this way need 10× less data, converge twice as fast, and perform better on both simulated benchmarks and real bimanual manipulation tasks. If robots can “imagine” motion using video, control becomes a much simpler problem. Shoutout to Jonas Pai, Liam Achenbach, Oier Mees, Elvis Nava and the rest of the team! Here's the project page: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

50,090 просмотров • 8 месяцев назад