Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

You’ve used language models, image models, video models, and voice models. Now it’s time for world models, thanks to World Labs.

Amir Haghighat

4,796 subscribers

839,697 views • 7 months ago •via X (Twitter)

Education News & Politics Science & Technology

Anya Rossi• Live Now

Private livecam show

30 Comments

Soren Iverson7 months ago

Finally, I can create a world where every interface is unhinged

World Labs7 months ago

🚀🌎

Amir Haghighat7 months ago

Honestly it's been awesome working with you all. @DanielChao4, @BenMildenhall, @drfeifei 💚

Philip Kiely7 months ago

World Labs is a vision of the future. My favorite thing to do in Marble is bring scenes from my favorite books to life and walk through them.

Charlie O'Neill7 months ago

Inference is everything everywhere all at once

Mudith Jayasekara7 months ago

a billion new worlds

Joaquin7 months ago

The "optic nerve" of AI is here. Simply epic. Massive h/t to the World Labs team for an incredible break through.

Rachel Rapp7 months ago

Insanely cool to work with pioneers in the field like @drfeifei and the entire team at World Labs 🩷

Mike Bilodeau7 months ago

a trillion new worlds!

ali7 months ago

can't wait to perf moggggggg

alana goyal7 months ago

so so cool

Madison Kanna7 months ago

excited!

Alina Weinstein7 months ago

a whole new world

James Weitzman7 months ago

🤯🤯🤯

Max Kirkby7 months ago

Everything is world

Aakash Gupta7 months ago

Keen to see the gaming studio uptake.

ALICE ⚡ | AI Growth Mentor7 months ago

Can't wait for this! OMG 🔥

Ian Curtis7 months ago

🌎

Tal Y7 months ago

wow 🤩🚀

Rahman.7 months ago

If you’ve mastered text, images, video, and audio, World Labs’ world models are the next level AI that understands reality itself

Sanskriti Sharma7 months ago

Whoa, super cool

Geet Khosla7 months ago

very cool

Vikrum Rathore7 months ago

This is platform power moving from language to lived digital space

Mark Jason7 months ago

that’s cool world models are the future @amiruci Let's DM

Vinay Rastogi7 months ago

The real paradigm shift is making worlds callable infrastructure 👏

Raees Ahmed7 months ago

Programmable 3D is where spatial AI stops being a demo and starts becoming a platform.

Prof 先生7 months ago

@DrClownPhD

Malav7 months ago

The state and pace of world models will actually define the timeline to AGI! This is a huge step towards the predictability for intelligence to infer the world context, its infinite subjectivity, and broadly perceptual intelligence with behavioral data at the base Way to go @amiruci @DanielChao4 @BenMildenhall @drfeifei 🔥

Jane Adams7 months ago

Because what this country really needed was more simulated reality from people who already live in one.

GaussianBrags7 months ago

You are not building world models though?

Related Videos

The next frontier after language models is world models, and learning to predict the next state of a world. Hear from our researchers how these models really work at a low-level and the mountain we're climbing. A general-purpose world simulator (AKA sci-fi) is now in sight.

The next frontier after language models is world models, and learning to predict the next state of a world. Hear from our researchers how these models really work at a low-level and the mountain we're climbing. A general-purpose world simulator (AKA sci-fi) is now in sight.

Odyssey

25,890 views • 10 months ago

better real-world physics are coming to video models thanks to synthetic video data

better real-world physics are coming to video models thanks to synthetic video data

Dreaming Tulpa 🥓👑

16,137 views • 1 year ago

Language models understand words. World models could help AI understand the world itself. TIME's Tharin Pillay spoke with Oliver Cameron, CEO of Odyssey, at AMD's Advancing AI event:

Language models understand words. World models could help AI understand the world itself. TIME's Tharin Pillay spoke with Oliver Cameron, CEO of Odyssey, at AMD's Advancing AI event:

TIME

49,444 views • 2 months ago

Persistent world models on the web. No 60-second time limit. World Labs × SparkJS × Three.js × Rapier 🌎

Persistent world models on the web. No 60-second time limit. World Labs × SparkJS × Three.js × Rapier 🌎

Ian Curtis

94,944 views • 8 months ago

Latent Spatial Memory for Video World Models

Latent Spatial Memory for Video World Models

AK

22,812 views • 3 months ago

It’s time to go beyond language models. Introducing Odyssey-2 Max, our most powerful world model yet. It materially advances the SOTA in physical accuracy. This is a big step toward models that simulate and interact with the world in real time. A new intelligence entirely!

It’s time to go beyond language models. Introducing Odyssey-2 Max, our most powerful world model yet. It materially advances the SOTA in physical accuracy. This is a big step toward models that simulate and interact with the world in real time. A new intelligence entirely!

Odyssey

2,187,596 views • 5 months ago

We’re giving away 1,000,000 simulations of reality, powered by our newest world model, Odyssey-2 Pro. Grab an API key, call /simulate, and get back a beautiful MP4. All free. The GPT-2 era rapidly explored language models. Now it’s time we explore world models together.

We’re giving away 1,000,000 simulations of reality, powered by our newest world model, Odyssey-2 Pro. Grab an API key, call /simulate, and get back a beautiful MP4. All free. The GPT-2 era rapidly explored language models. Now it’s time we explore world models together.

Odyssey

40,935 views • 8 months ago

My response to my opinion on World of Outlaws Late Models and Lucas Oil late models 🏁

My response to my opinion on World of Outlaws Late Models and Lucas Oil late models 🏁

Kenny Wallace

28,639 views • 2 years ago

Large language models reason through text. Vision‑language‑action models reason through the real world. By fusing perception, context, and action from live video, VLAs deliver the awareness physical AI needs for next‑gen robotics and edge systems.

Large language models reason through text. Vision‑language‑action models reason through the real world. By fusing perception, context, and action from live video, VLAs deliver the awareness physical AI needs for next‑gen robotics and edge systems.

Intel

15,931 views • 7 months ago

MindJourney Test-Time Scaling with World Models for Spatial Reasoning

MindJourney Test-Time Scaling with World Models for Spatial Reasoning

AK

61,826 views • 1 year ago

Why LLMs are a dead end for human-level intelligence, and especially for Physical AI / Robotics. The next leap isn’t bigger language models. It’s World Models. I just dropped a full 1-hour presentation from Shanghai: “World Models: the ChatGPT moment for robotics?” → Why LLMs hit a wall → Why action-conditioned world models planning in latent space are the real path → Live World Forge demo with LeWorldModel + Hugging Face LeRobot Watch here. The future of intelligence is embodied, not just chatty.

Why LLMs are a dead end for human-level intelligence, and especially for Physical AI / Robotics. The next leap isn’t bigger language models. It’s World Models. I just dropped a full 1-hour presentation from Shanghai: “World Models: the ChatGPT moment for robotics?” → Why LLMs hit a wall → Why action-conditioned world models planning in latent space are the real path → Live World Forge demo with LeWorldModel + Hugging Face LeRobot Watch here. The future of intelligence is embodied, not just chatty.

abdel

39,459 views • 4 months ago

Can we extend the power of world models beyond just online model-based learning? Absolutely! We believe the true potential of world models lies in enabling agents to reason at test time. Introducing DINO-WM: World Models on Pre-trained Visual Features for Zero-shot Planning.

Can we extend the power of world models beyond just online model-based learning? Absolutely! We believe the true potential of world models lies in enabling agents to reason at test time. Introducing DINO-WM: World Models on Pre-trained Visual Features for Zero-shot Planning.

Gaoyue Zhou

168,305 views • 1 year ago

.Oliver Cameron says one of the biggest applications for world models will be training AI inside simulated environments. "One of the killer applications of world models will be as a learning environment where language models and other types of AI can learn. It gets very much philosophical."

.Oliver Cameron says one of the biggest applications for world models will be training AI inside simulated environments. "One of the killer applications of world models will be as a learning environment where language models and other types of AI can learn. It gets very much philosophical."

TBPN

40,172 views • 2 months ago

The only path forward is with video and world models.

The only path forward is with video and world models.

Cristóbal Valenzuela

21,621 views • 9 months ago

[New post + market map + video] 🔥 AI Learned to Talk. Now it’s Learning to Build Reality What if you could step into any world you can imagine? Not just view it, but walk around, interact, and create within it? That's the promise of AI World Models. The leap from language models to world models is like going from reading about Paris to actually walking its streets. This isn't incremental—it's transformative for how we'll experience digital content, do our jobs, and interact with technology.

[New post + market map + video] 🔥 AI Learned to Talk. Now it’s Learning to Build Reality What if you could step into any world you can imagine? Not just view it, but walk around, interact, and create within it? That's the promise of AI World Models. The leap from language models to world models is like going from reading about Paris to actually walking its streets. This isn't incremental—it's transformative for how we'll experience digital content, do our jobs, and interact with technology.

Colin Campbell

44,630 views • 1 year ago

Hedra MCP + CLI are live: the first step toward our platform for building, improving, and deploying vision models. Connect Claude Code, ChatGPT, Hermes, or your own agents to Hedra's image, video, and world models, right where you already work. Docs 👇

Hedra MCP + CLI are live: the first step toward our platform for building, improving, and deploying vision models. Connect Claude Code, ChatGPT, Hermes, or your own agents to Hedra's image, video, and world models, right where you already work. Docs 👇

Hedra

88,199 views • 5 days ago

Duracell spec ad made completely on invideo Agents & Models. What is Agents & Models? Access the world's best image, video, and audio models in a dedicated environment designed for control and end-to-end project management. Take complete control of your creative output. Get access to all image models with unlimited usage on all paid plans for the next 365 days.

Duracell spec ad made completely on invideo Agents & Models. What is Agents & Models? Access the world's best image, video, and audio models in a dedicated environment designed for control and end-to-end project management. Take complete control of your creative output. Get access to all image models with unlimited usage on all paid plans for the next 365 days.

Invideo

437,527 views • 8 months ago

UpliftAI ( develops Voice Models for regional languages like Urdu and Greek. 60% of the world only speaks a regional language, and current models don't work. UpliftAI found a way to gather data for languages at unprecedented scale. Congrats on the launch, Hammad Malik, Zaid & Muhammad Bin Sabir!

UpliftAI ( develops Voice Models for regional languages like Urdu and Greek. 60% of the world only speaks a regional language, and current models don't work. UpliftAI found a way to gather data for languages at unprecedented scale. Congrats on the launch, Hammad Malik, Zaid & Muhammad Bin Sabir!

Y Combinator

21,091 views • 1 year ago