Загрузка видео...

Не удалось загрузить видео

Возникла проблема при загрузке этого видео. Это может быть связано с временными проблемами сети или видео может быть недоступно.

На главную

What if we can take a few photos and turn it into an interactive physically realistic virtual world🌎? 📢Introducing AdaVoMP (ICML 26): generates volumetric physics fields at high spatial resolution making objects interactive and deformable🦾 🦣16^3x higher res (1024^3) ⚡️more accurate

rishit dagli

4,571 subscribers

14,908 просмотров • 1 месяц назад •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

What if point particles in Physics are just an illusion? Branes are higher-dimensional objects in String Theory that generalize the idea of a point particle. Instead of everything being just 0-dimensional points or 1-dimensional strings, a Brane can have 2, 3 or more spatial dimensions Open strings can end on Branes, which means Branes act like boundaries for string motion. In many models, what we call particles are simply vibration modes of strings attached to these Branes, and force can be understood as strings stretching between them.

What if point particles in Physics are just an illusion? Branes are higher-dimensional objects in String Theory that generalize the idea of a point particle. Instead of everything being just 0-dimensional points or 1-dimensional strings, a Brane can have 2, 3 or more spatial dimensions Open strings can end on Branes, which means Branes act like boundaries for string motion. In many models, what we call particles are simply vibration modes of strings attached to these Branes, and force can be understood as strings stretching between them.

Mathelirium

21,841 просмотров • 2 месяцев назад

Another city. Another full 3D Gaussian Splat reconstruction. This entire neighborhood was captured while walking with an Insta360 X5, then processed with our GPU pipeline and transformed into a high-fidelity, machine-readable 3D environment with Over the Reality. What you’re seeing is not an isolated virtual world. It’s a 3D digital twin of a real-world location, built from structured spatial data where every building, street, and detail can become usable by AI, robotics, and XR. And it’s happening at scale, across cities in more than 180 countries every day. Have an Insta360 X5 and want to become a mapper? Apply here: Want to build a 3D Gaussian Splat from a video you already recorded? Do it for free at powered by Over the Reality.

Another city. Another full 3D Gaussian Splat reconstruction. This entire neighborhood was captured while walking with an Insta360 X5, then processed with our GPU pipeline and transformed into a high-fidelity, machine-readable 3D environment with Over the Reality. What you’re seeing is not an isolated virtual world. It’s a 3D digital twin of a real-world location, built from structured spatial data where every building, street, and detail can become usable by AI, robotics, and XR. And it’s happening at scale, across cities in more than 180 countries every day. Have an Insta360 X5 and want to become a mapper? Apply here: Want to build a 3D Gaussian Splat from a video you already recorded? Do it for free at powered by Over the Reality.

Over the Reality 🌐

400,097 просмотров • 3 месяцев назад

These three IIT Madras scientists are insane. Their startup Sthyr Energy is literally turning electricity into metal. Which you can keep for months and turn it back into electricity when you need it. And this is incredibly huge. Let's break this down. Right now, India alone generates enough renewable energy to power countries like France. But we can either use it as soon as its generated or its lost forever. Because no one has figured out a way to store electricity for more than a few hours at scale. If Sthyr's solution works - we won't just be able to store it for years but we could also transport it on roads - without creating any new infrastructure. And it would change how the world uses electricity forever.

These three IIT Madras scientists are insane. Their startup Sthyr Energy is literally turning electricity into metal. Which you can keep for months and turn it back into electricity when you need it. And this is incredibly huge. Let's break this down. Right now, India alone generates enough renewable energy to power countries like France. But we can either use it as soon as its generated or its lost forever. Because no one has figured out a way to store electricity for more than a few hours at scale. If Sthyr's solution works - we won't just be able to store it for years but we could also transport it on roads - without creating any new infrastructure. And it would change how the world uses electricity forever.

Varun Guru

181,957 просмотров • 3 месяцев назад

We releasing an early version of Moonlake's world modeling agent. It is not just a tool for generating beautiful scenes. It is a system for creating worlds you can interact with. Objects respond. Physics runs determinstically. Your actions have real causal effects. We believe in building with the community, not in isolation. Over the past month, we have iterated closely with users, hosted hackathons, and worked side by side with hundreds of beginner game developers to understand their workflows, frustrations, and creative ambitions firsthand. Our intention is simple. Help developers solve real pain points. Expensive tooling. Repetitive tasks. Tedious work that slow down creativity. We believe humans are the true source of creativity. Tools should amplify that power, not replace it. Game developers, and more broadly creators, will always be at the center of our ecosystem. This is our first beta release. It is an early step, but we hope it gives you a clear sense of what we are building toward. A future where anyone can create rich, interactive worlds with intention and meaning.

We releasing an early version of Moonlake's world modeling agent. It is not just a tool for generating beautiful scenes. It is a system for creating worlds you can interact with. Objects respond. Physics runs determinstically. Your actions have real causal effects. We believe in building with the community, not in isolation. Over the past month, we have iterated closely with users, hosted hackathons, and worked side by side with hundreds of beginner game developers to understand their workflows, frustrations, and creative ambitions firsthand. Our intention is simple. Help developers solve real pain points. Expensive tooling. Repetitive tasks. Tedious work that slow down creativity. We believe humans are the true source of creativity. Tools should amplify that power, not replace it. Game developers, and more broadly creators, will always be at the center of our ecosystem. This is our first beta release. It is an early step, but we hope it gives you a clear sense of what we are building toward. A future where anyone can create rich, interactive worlds with intention and meaning.

Fan-Yun Sun

304,827 просмотров • 5 месяцев назад

Introducing /visual-plan - a skill to generate rich, visual plans for Claude Code and Codex. Plan mode in Claude Code is incredible. But I always find my eyes glazing over when it gives me this huge markdown essay in my terminal. I found I can make much better visual plans with reusable components. So I made a skill called `/visual-plan`. It generates plans as MDX with visual, interactive components. Diagrams, interactive API specs, schema design changes, annotated code, and even pan and zoomable wireframes. So for any UI work, you can look at a wireframe first, comment on it, iterate, and then have the agent work. I’ve found this to be a much more intuitive interface for reasoning about what the agent is doing. It’s somewhat inspired by that popular post about how HTML is better than Markdown. But HTML can be slow and verbose to write. And it doesn’t look good checked into a repo. This has really made me feel like humans and engineering are entering a new abstraction phase, where we reason about things at the plan level. As long as the plan is good, agents are getting more and more reliable at executing on it. Almost to the degree that we trust the C compiler to compile to assembly reliably. Plans are the new intermediate representation. I also made a skill for the reverse of this, called `/visual-recap`. After the agent works, it gives you a recap of everything it did. Same idea: wireframes, interactive API specs and diffs, schemas, annotated code, etc. So now when you’re reviewing what the agent did for you, or looking at a pull request of somebody else’s code, you can see a visual recap instead of just reading a wall of text. It’s all free and open source. You can find it on my GitHub. Will link to it in the reply because we all know how dumb these algorithms are with links.

Introducing /visual-plan - a skill to generate rich, visual plans for Claude Code and Codex. Plan mode in Claude Code is incredible. But I always find my eyes glazing over when it gives me this huge markdown essay in my terminal. I found I can make much better visual plans with reusable components. So I made a skill called `/visual-plan`. It generates plans as MDX with visual, interactive components. Diagrams, interactive API specs, schema design changes, annotated code, and even pan and zoomable wireframes. So for any UI work, you can look at a wireframe first, comment on it, iterate, and then have the agent work. I’ve found this to be a much more intuitive interface for reasoning about what the agent is doing. It’s somewhat inspired by that popular post about how HTML is better than Markdown. But HTML can be slow and verbose to write. And it doesn’t look good checked into a repo. This has really made me feel like humans and engineering are entering a new abstraction phase, where we reason about things at the plan level. As long as the plan is good, agents are getting more and more reliable at executing on it. Almost to the degree that we trust the C compiler to compile to assembly reliably. Plans are the new intermediate representation. I also made a skill for the reverse of this, called `/visual-recap`. After the agent works, it gives you a recap of everything it did. Same idea: wireframes, interactive API specs and diffs, schemas, annotated code, etc. So now when you’re reviewing what the agent did for you, or looking at a pull request of somebody else’s code, you can see a visual recap instead of just reading a wall of text. It’s all free and open source. You can find it on my GitHub. Will link to it in the reply because we all know how dumb these algorithms are with links.

Steve (Builder.io)

124,409 просмотров • 1 месяц назад

Today, we're joined by Jack Parker-Holder and Shlomi Fruchter, researchers at Google DeepMind, to discuss the recent release of Genie 3, a model capable of generating “playable” virtual worlds. We dig into the evolution of the Genie project and review the current model’s scaled-up capabilities, including creating real-time, interactive, and high-resolution environments. Jack and Shlomi share their perspectives on what defines a world model, the model's architecture, and key technical challenges and breakthroughs, including Genie 3’s visual memory and ability to handle “promptable world events.” Jack, Shlomi, and Sam share their favorite Genie 3 demos, and discuss its potential as a dynamic training environment for embodied AI agents. Finally, we will explore future directions for Genie research. 🗒️ For the full list of resources for this episode, visit the show notes page: 📖 CHAPTERS =============================== 00:00 - Introduction 7:11 - What is a world model? 14:49 - Milestones of Genie research 24:32 - Genie 3 27:46 - Challenges 30:07 - Genie 3 examples 33:48 - Model capabilities 35:49 - Key aspects of the model 39:40 - Consistency as an emergent property 42:11 - Promptable word events 47:24 - SIMA agent 50:56 - Limitations 56:08 - Future directions

Today, we're joined by Jack Parker-Holder and Shlomi Fruchter, researchers at Google DeepMind, to discuss the recent release of Genie 3, a model capable of generating “playable” virtual worlds. We dig into the evolution of the Genie project and review the current model’s scaled-up capabilities, including creating real-time, interactive, and high-resolution environments. Jack and Shlomi share their perspectives on what defines a world model, the model's architecture, and key technical challenges and breakthroughs, including Genie 3’s visual memory and ability to handle “promptable world events.” Jack, Shlomi, and Sam share their favorite Genie 3 demos, and discuss its potential as a dynamic training environment for embodied AI agents. Finally, we will explore future directions for Genie research. 🗒️ For the full list of resources for this episode, visit the show notes page: 📖 CHAPTERS =============================== 00:00 - Introduction 7:11 - What is a world model? 14:49 - Milestones of Genie research 24:32 - Genie 3 27:46 - Challenges 30:07 - Genie 3 examples 33:48 - Model capabilities 35:49 - Key aspects of the model 39:40 - Consistency as an emergent property 42:11 - Promptable word events 47:24 - SIMA agent 50:56 - Limitations 56:08 - Future directions

The TWIML AI Podcast

12,043 просмотров • 11 месяцев назад

here's what we built at the Sarvam x GrowthX hackathon! > an interactive, voice-first game where you learn new languages through everyday conversations > our insight was that not everyone needs to master a language - they only need a few sentences, useful in everyday situations > you can choose your region (like Bengaluru), and the level of difficulty according to which your interactions are based > it then guides you on what to say in a given situation, like booking an auto > it grades on you how well you spoke, what you missed, progressively getting difficult with more levels didn't make it to the top 15, but hey we had fun :) Sarvam for Developers's TTS-STT stack is definitely some of the best in the world currently.

here's what we built at the Sarvam x GrowthX hackathon! > an interactive, voice-first game where you learn new languages through everyday conversations > our insight was that not everyone needs to master a language - they only need a few sentences, useful in everyday situations > you can choose your region (like Bengaluru), and the level of difficulty according to which your interactions are based > it then guides you on what to say in a given situation, like booking an auto > it grades on you how well you spoke, what you missed, progressively getting difficult with more levels didn't make it to the top 15, but hey we had fun :) Sarvam for Developers's TTS-STT stack is definitely some of the best in the world currently.

Parth

40,152 просмотров • 3 дней назад

🚀 We introduce Neural Theorizer (NEO) — a new type of world model that learns to theorize the world from observation, without language or LLM supervision. Selected as an ICML 2026 oral presentation — 0.7% of submitted papers. The paper asks: "What does it mean to understand the world and build a world model?" Today’s world models are often trained to predict the future: the next frame, next latent state, or next observation. But is prediction enough? We argue that a world model should be a theory-building system: one that discovers reusable primitives, composes them into executable explanations, and transfers those explanations to novel phenomena. NEO is our first step toward this vision — a World Theory Model that learns explicit, compositional theories from raw observation. This work was led by my wonderful students: Doojin Baek*(Doojin Baek), Gyubin Lee* (GyuBin Lee), Junyeob Baek (Junyeob Baek), and Hosung Lee (Hosung Lee). For more details, take a look at the paper — and if you’re attending ICML, let’s talk there! 📄 arXiv: 🌐 Project page:

🚀 We introduce Neural Theorizer (NEO) — a new type of world model that learns to theorize the world from observation, without language or LLM supervision. Selected as an ICML 2026 oral presentation — 0.7% of submitted papers. The paper asks: "What does it mean to understand the world and build a world model?" Today’s world models are often trained to predict the future: the next frame, next latent state, or next observation. But is prediction enough? We argue that a world model should be a theory-building system: one that discovers reusable primitives, composes them into executable explanations, and transfers those explanations to novel phenomena. NEO is our first step toward this vision — a World Theory Model that learns explicit, compositional theories from raw observation. This work was led by my wonderful students: Doojin Baek(Doojin Baek), Gyubin Lee (GyuBin Lee), Junyeob Baek (Junyeob Baek), and Hosung Lee (Hosung Lee). For more details, take a look at the paper — and if you’re attending ICML, let’s talk there! 📄 arXiv: 🌐 Project page:

Sungjin Ahn

98,596 просмотров • 1 месяц назад

Dormant Memories — a gallery regrouping my recent experiments built from real-world captures, reconstructed as 3D Gaussian Splats, and in many cases manipulated using World Labs' Marble. Everything is built with Spark, the 3D Gaussian splat renderer released by World Labs. It runs in the browser and can be experienced across devices: desktop, mobile, and headsets like Apple Vision Pro and Quest 3. This work comes out of something I've been exploring for a few years now: capturing fragments of the world (places, objects, moments...) and turning them into spaces that can then be revisited, manipulated, and explored in real time. What particularly interests me here is the process of creating alternate versions of the same place; starting from something that physically exists and using it as a base to explore variations, instead of generating from scratch. These explorations are part of what we do at Dpt.... same tools, same curiosity, same collaborative energy that shape the studio's work on interactive and immersive experiences. Some of these experiments were featured as part of the Spark 2.0 release by World Labs. Their technical blog post is a must-read; link in the thread ↓ These are experiments. They aren't fully optimized, they aren't bug-free, and the UX has rough edges. The goal is to explore, test, and learn. #GaussianSplatting #3DGS #pointcloud #WorldModels #sparkjs #threejs

Dormant Memories — a gallery regrouping my recent experiments built from real-world captures, reconstructed as 3D Gaussian Splats, and in many cases manipulated using World Labs' Marble. Everything is built with Spark, the 3D Gaussian splat renderer released by World Labs. It runs in the browser and can be experienced across devices: desktop, mobile, and headsets like Apple Vision Pro and Quest 3. This work comes out of something I've been exploring for a few years now: capturing fragments of the world (places, objects, moments...) and turning them into spaces that can then be revisited, manipulated, and explored in real time. What particularly interests me here is the process of creating alternate versions of the same place; starting from something that physically exists and using it as a base to explore variations, instead of generating from scratch. These explorations are part of what we do at Dpt.... same tools, same curiosity, same collaborative energy that shape the studio's work on interactive and immersive experiences. Some of these experiments were featured as part of the Spark 2.0 release by World Labs. Their technical blog post is a must-read; link in the thread ↓ These are experiments. They aren't fully optimized, they aren't bug-free, and the UX has rough edges. The goal is to explore, test, and learn. #GaussianSplatting #3DGS #pointcloud #WorldModels #sparkjs #threejs

Hugues Bruyère

12,408 просмотров • 3 месяцев назад

🚀 Announcing Echo — our new frontier model for 3D world generation. Echo turns a simple text prompt or image into a fully explorable, 3D-consistent world. Instead of disconnected views, the result is a single, coherent spatial representation you can move through freely. This is part of a bigger shift in AI: from generating pixels and tokens to generating spaces. Echo predicts a geometry-grounded 3D scene at metric scale, meaning every novel view, depth map, and interaction comes from the same underlying world — not independent hallucinations. Once generated, the world is interactive in real time. You control the camera, explore from any angle, and render instantly — even on low-end hardware, directly in the browser. High-quality 3D world exploration is no longer gated by expensive equipment. Under the hood, Echo infers a physically grounded 3D representation and converts it into a renderable format. For our web demo, we use 3D Gaussian Splatting (3DGS) for fast, GPU-friendly rendering — but the representation itself is flexible and can be easily adapted. Why this matters: consistent 3D worlds unlock real workflows — digital twins, 3D design, game environments, robotics simulation, and more. From a single photo or a line of text, Echo builds worlds that are reliable, editable, and spatially faithful. Echo also enables scene editing and restyling. Change materials, remove or add objects, explore design variations — all while preserving global 3D consistency. Editing no longer breaks the world. This is only the beginning. Echo is the foundation for future world models with dynamics, physical reasoning, and richer interaction — environments that don’t just look right, but behave right. Explore the generated worlds on our website and sign up for the closed beta. The era of spatial intelligence starts here. 🌍 #Echo #WorldModels #SpatialAI #3DFoundationModels Check it out:

🚀 Announcing Echo — our new frontier model for 3D world generation. Echo turns a simple text prompt or image into a fully explorable, 3D-consistent world. Instead of disconnected views, the result is a single, coherent spatial representation you can move through freely. This is part of a bigger shift in AI: from generating pixels and tokens to generating spaces. Echo predicts a geometry-grounded 3D scene at metric scale, meaning every novel view, depth map, and interaction comes from the same underlying world — not independent hallucinations. Once generated, the world is interactive in real time. You control the camera, explore from any angle, and render instantly — even on low-end hardware, directly in the browser. High-quality 3D world exploration is no longer gated by expensive equipment. Under the hood, Echo infers a physically grounded 3D representation and converts it into a renderable format. For our web demo, we use 3D Gaussian Splatting (3DGS) for fast, GPU-friendly rendering — but the representation itself is flexible and can be easily adapted. Why this matters: consistent 3D worlds unlock real workflows — digital twins, 3D design, game environments, robotics simulation, and more. From a single photo or a line of text, Echo builds worlds that are reliable, editable, and spatially faithful. Echo also enables scene editing and restyling. Change materials, remove or add objects, explore design variations — all while preserving global 3D consistency. Editing no longer breaks the world. This is only the beginning. Echo is the foundation for future world models with dynamics, physical reasoning, and richer interaction — environments that don’t just look right, but behave right. Explore the generated worlds on our website and sign up for the closed beta. The era of spatial intelligence starts here. 🌍 #Echo #WorldModels #SpatialAI #3DFoundationModels Check it out:

SpAItial AI

176,105 просмотров • 7 месяцев назад

ELON: TUNNELS CAN SOLVE TRAFFIC AND TURN CITIES INTO PARKS “The Boring Company is making good progress. For the longest time, people would ask me what opportunities exist, and I would say tunnels. If someone could just do a tunneling company, we could solve traffic and also have very high speed transport between cities. I am pro train. I love trains. But trains cut through towns, split property, and are very loud. A simplified Hyperloop is really just Teslas in a depressurized tunnel. You can go between cities super fast without disrupting anything above ground. The vision is to make roads three dimensional with multiple layers of tunnels. You can solve essentially any traffic problem in any city if you go 3D. You can even turn streets into parks because you will not need cars or parking. We already have an operational tunnel in Vegas. It can transform cities far more than people realize.” Source: Eva Fo𝕏 🦊 Claudius Nero's Legion Elon Musk

ELON: TUNNELS CAN SOLVE TRAFFIC AND TURN CITIES INTO PARKS “The Boring Company is making good progress. For the longest time, people would ask me what opportunities exist, and I would say tunnels. If someone could just do a tunneling company, we could solve traffic and also have very high speed transport between cities. I am pro train. I love trains. But trains cut through towns, split property, and are very loud. A simplified Hyperloop is really just Teslas in a depressurized tunnel. You can go between cities super fast without disrupting anything above ground. The vision is to make roads three dimensional with multiple layers of tunnels. You can solve essentially any traffic problem in any city if you go 3D. You can even turn streets into parks because you will not need cars or parking. We already have an operational tunnel in Vegas. It can transform cities far more than people realize.” Source: Eva Fo𝕏 🦊 Claudius Nero's Legion Elon Musk

Mario Nawfal

15,685 просмотров • 7 месяцев назад

Robotics has a massive, silent bottleneck. It isn’t just data collection—it’s the brutal 1x speed of the physical world. Genesis AI Genesis AI just unveiled Genesis World 1.0, and they are attempting to turn the notorious Sim2Real gap into a pure compute problem. Evaluating a robotics foundation model across edge cases usually means hundreds of hours of physical lab testing. With Genesis World 1.0, what traditionally takes nearly a week of continuous, real-world operation is being compressed into 30 minutes in simulation. What makes this different from just dropping a robot model into an off-the-shelf game engine? 1️⃣ Nyx Renderer: A custom, real-time path-traced engine rendering noise-free 1080p frames in under 4ms. Game engines use rasterization tricks that confuse AI; Nyx uses physically accurate multi-bounce lighting so the model's "eyes" see exactly what real sensors see. 2️⃣ Quadrants Compiler: A custom Python-to-GPU compiler to run heavily parallelized multi-physics simulations (rigid bodies, fluids, deformables) natively across architectures. 3️⃣ Evaluation First: They aren't rushing to train on synthetic data. They are using this purely for closed-loop evaluation to perfect the physics first, currently claiming an impressive 89% correlation with real-world hardware tests. If the industry can accurately evaluate models in simulation without the physical world bottleneck, humanoid development stops moving at wall-clock time and starts scaling with compute.

Robotics has a massive, silent bottleneck. It isn’t just data collection—it’s the brutal 1x speed of the physical world. Genesis AI Genesis AI just unveiled Genesis World 1.0, and they are attempting to turn the notorious Sim2Real gap into a pure compute problem. Evaluating a robotics foundation model across edge cases usually means hundreds of hours of physical lab testing. With Genesis World 1.0, what traditionally takes nearly a week of continuous, real-world operation is being compressed into 30 minutes in simulation. What makes this different from just dropping a robot model into an off-the-shelf game engine? 1️⃣ Nyx Renderer: A custom, real-time path-traced engine rendering noise-free 1080p frames in under 4ms. Game engines use rasterization tricks that confuse AI; Nyx uses physically accurate multi-bounce lighting so the model's "eyes" see exactly what real sensors see. 2️⃣ Quadrants Compiler: A custom Python-to-GPU compiler to run heavily parallelized multi-physics simulations (rigid bodies, fluids, deformables) natively across architectures. 3️⃣ Evaluation First: They aren't rushing to train on synthetic data. They are using this purely for closed-loop evaluation to perfect the physics first, currently claiming an impressive 89% correlation with real-world hardware tests. If the industry can accurately evaluate models in simulation without the physical world bottleneck, humanoid development stops moving at wall-clock time and starts scaling with compute.

Humanoids daily

17,240 просмотров • 2 месяцев назад

Everything you love about generative models — now powered by real physics! Announcing the Genesis project — after a 24-month large-scale research collaboration involving over 20 research labs — a generative physics engine able to generate 4D dynamical worlds powered by a physics simulation platform designed for general-purpose robotics and physical AI applications. Genesis's physics engine is developed in pure Python, while being 10-80x faster than existing GPU-accelerated stacks like Isaac Gym and MJX. It delivers a simulation speed ~430,000 faster than in real-time, and takes only 26 seconds to train a robotic locomotion policy transferrable to the real world on a single RTX4090 (see tutorial: The Genesis physics engine and simulation platform is fully open source at We'll gradually roll out access to our generative framework in the near future. Genesis implements a unified simulation framework all from scratch, integrating a wide spectrum of state-of-the-art physics solvers, allowing simulation of the whole physical world in a virtual realm with the highest realism. We aim to build a universal data engine that leverages an upper-level generative framework to autonomously create physical worlds, together with various modes of data, including environments, camera motions, robotic task proposals, reward functions, robot policies, character motions, fully interactive 3D scenes, open-world articulated assets, and more, aiming towards fully automated data generation for robotics, physical AI and other applications. Open Source Code: Project webpage: Documentation: 1/n

Everything you love about generative models — now powered by real physics! Announcing the Genesis project — after a 24-month large-scale research collaboration involving over 20 research labs — a generative physics engine able to generate 4D dynamical worlds powered by a physics simulation platform designed for general-purpose robotics and physical AI applications. Genesis's physics engine is developed in pure Python, while being 10-80x faster than existing GPU-accelerated stacks like Isaac Gym and MJX. It delivers a simulation speed ~430,000 faster than in real-time, and takes only 26 seconds to train a robotic locomotion policy transferrable to the real world on a single RTX4090 (see tutorial: The Genesis physics engine and simulation platform is fully open source at We'll gradually roll out access to our generative framework in the near future. Genesis implements a unified simulation framework all from scratch, integrating a wide spectrum of state-of-the-art physics solvers, allowing simulation of the whole physical world in a virtual realm with the highest realism. We aim to build a universal data engine that leverages an upper-level generative framework to autonomously create physical worlds, together with various modes of data, including environments, camera motions, robotic task proposals, reward functions, robot policies, character motions, fully interactive 3D scenes, open-world articulated assets, and more, aiming towards fully automated data generation for robotics, physical AI and other applications. Open Source Code: Project webpage: Documentation: 1/n

Zhou Xian

3,818,087 просмотров • 1 год назад

In 2026, AI world models will take the spotlight in storytelling - powering new types of interactive experiences & digital economies not seen before World models are progressing rapidly - Marble World Labs and Genie 3 Google DeepMind already generate 3D environments from text prompts, allowing users to explore them as if they were video games As creators adopt these tools, new storytelling formats will emerge. One genre I'm excited about is "generative Minecraft" - where players co-create virtual worlds together by vibe coding with world models. Game mechanics could be programmable with natural language - ex. "create a paintbrush that changes the color of anything I touch to pink" World models will also likely give rise to not just a single game, but an entire new category of generative world experiences - you could have a horror experience where you’re hiding from generated monsters, or a D&D experience where you’re roaming an infinite fantasy world with friends And with a common base model for the underlying worlds, these experiences could be inter-connected in a multiverse we could only dream about previously in science fiction A key affordance here is the role of consumers as co-creators - you can wander the multiverse as a tourist, or break out your pickaxe and become a creator anytime. This in turn would give rise to new digital economies - with creators making a living building and selling interoperable assets, serving as a guide for new players, etc The opportunity is enormous - a new category of generative worlds would not only create a new storytelling medium unlike any we’ve seen before, but also be rich training grounds for agents, robotics, and AGI If you’re excited about building new interactive experiences or the virtual economy stack with world models - we’d love to hear from you!

In 2026, AI world models will take the spotlight in storytelling - powering new types of interactive experiences & digital economies not seen before World models are progressing rapidly - Marble World Labs and Genie 3 Google DeepMind already generate 3D environments from text prompts, allowing users to explore them as if they were video games As creators adopt these tools, new storytelling formats will emerge. One genre I'm excited about is "generative Minecraft" - where players co-create virtual worlds together by vibe coding with world models. Game mechanics could be programmable with natural language - ex. "create a paintbrush that changes the color of anything I touch to pink" World models will also likely give rise to not just a single game, but an entire new category of generative world experiences - you could have a horror experience where you’re hiding from generated monsters, or a D&D experience where you’re roaming an infinite fantasy world with friends And with a common base model for the underlying worlds, these experiences could be inter-connected in a multiverse we could only dream about previously in science fiction A key affordance here is the role of consumers as co-creators - you can wander the multiverse as a tourist, or break out your pickaxe and become a creator anytime. This in turn would give rise to new digital economies - with creators making a living building and selling interoperable assets, serving as a guide for new players, etc The opportunity is enormous - a new category of generative worlds would not only create a new storytelling medium unlike any we’ve seen before, but also be rich training grounds for agents, robotics, and AGI If you’re excited about building new interactive experiences or the virtual economy stack with world models - we’d love to hear from you!

Jon Lai

29,871 просмотров • 7 месяцев назад

Fast Company just published a great piece on World Labs , Fei-Fei Li , Marble, and the idea that spatial intelligence / world models may be one of the next big shifts in AI. I was happy to be quoted in the article, but I also wanted to share more context about my own experience with World Labs and Marble, and why this direction is especially interesting to me. My starting point: volumetric capture — For the past few years I’ve been exploring and using volumetric capture and reconstruction (photogrammetry, NeRFs, 3D Gaussian Splats) mostly capturing locations around Montreal. Alleys, museums, urban interiors. I love every step of it: the capture itself, the pipeline, and what can be done with the output. Turning real spaces into real-time explorable systems. I do this personally, sharing explorations here, and professionally as chief technologist, and co-founder of Dpt. Physical reality + generative manipulation — In my work I’m especially drawn to mixing physical reality with generative and digital manipulation: using physical interfaces (light, clay, ink, ... ) to drive generative AI pipelines, building mixed reality prototypes that reshape your surroundings, or starting from real captured spaces and transforming them using tools like Marble. Like many people, I saw the World Labs announcement on Twitter in September 2024, and Marble when it surfaced in early December. But by then, I already had a sense something was coming. The first conversation — As someone deep into volumetric capture and radiance fields, I obviously knew about Ben Mildenhall and his pioneering work on NeRF. To my surprise, Ben reached out to me in late June 2024. He’d been following some of my experiments and wanted to chat about my process and workflows and how I was using this “stuff” creatively. At that point he didn’t share what he was building, but we had a genuinely great conversation about radiance fields, AI, and my work. He was curious about the creative perspective, not just the technical one. When the World Labs announcement dropped a few months later, it all made sense. I understood what Ben had been working on, and why the creative angle mattered to them. Then in August 2025, he invited me to try the Marble beta, and I’ve been experimenting with it since. Experimenting with Marble — The first thing I used Marble for was materializing scene and world concepts during ideation at the studio, and seeing if and how it could fit into our production pipeline. In parallel, I dove into a series of experiments focused on world manipulation: starting from real captured spaces and transforming them using Marble. I’d already been exploring that idea using img2img diffusion with ControlNet on NeRF renders, real-time video streams, and even mixed reality using headset camera feeds. But Marble brings something different. It generates persistent, spatially cohesive 3D worlds that can be rendered in real time across a wide range of devices. That’s a real shift. Experiment 01: Parallel Realities — The first experiment, Parallel Realities, starts from a volumetric capture of a real location, reconstructed as 3D Gaussian Splats. Using Marble, I generate an alternate version of that same space, something informed by the original architecture: abandoned, nature-reclaimed, alternate era. Then, using Spark (World Labs’ 3D Gaussian Splatting renderer for THREE.js) I make both realities coexist in the same spatial coordinate system. From there, I use a portal UX mechanic to let the user step between the real reconstruction and the Marble-generated version. Experiment 02: Hidden Depth The second experiment, Hidden Depth, does not transform a space as much as expand it. A captured location has a visual boundary (a mural, a doorway, a dark corridor) and Marble generates what exists beyond it. For example: a Montreal alley has a painted mural; step through it and you’re inside a world informed by what is actually depicted there. World Labs showcased part of this work here: And in their Spark 2.0 post: The project page is here: Why this matters to me — Being able to start from a real 3D Gaussian Splat scene and manipulate it with Marble opens up a lot of ideas. The 3DGS pipeline is becoming an increasingly compelling foundation for exploration, experimentation, and storytelling. What matters most to me right now is more control. The more I can steer the generated scene or world, the more useful the tool becomes. I want more features like the already existing multiple input images and Chisel, the blockout-based approach. I would like better local control, the ability to expand a generated world more and more while preserving coherence, and the ability to directly import 3D Gaussian Splat scenes to be used as a starting point. I want more ways to shape the result, not just a “prompt and hope” approach. — It is exciting to see this field moving from research and demos toward actual creative workflows.

Fast Company just published a great piece on World Labs , Fei-Fei Li , Marble, and the idea that spatial intelligence / world models may be one of the next big shifts in AI. I was happy to be quoted in the article, but I also wanted to share more context about my own experience with World Labs and Marble, and why this direction is especially interesting to me. My starting point: volumetric capture — For the past few years I’ve been exploring and using volumetric capture and reconstruction (photogrammetry, NeRFs, 3D Gaussian Splats) mostly capturing locations around Montreal. Alleys, museums, urban interiors. I love every step of it: the capture itself, the pipeline, and what can be done with the output. Turning real spaces into real-time explorable systems. I do this personally, sharing explorations here, and professionally as chief technologist, and co-founder of Dpt. Physical reality + generative manipulation — In my work I’m especially drawn to mixing physical reality with generative and digital manipulation: using physical interfaces (light, clay, ink, ... ) to drive generative AI pipelines, building mixed reality prototypes that reshape your surroundings, or starting from real captured spaces and transforming them using tools like Marble. Like many people, I saw the World Labs announcement on Twitter in September 2024, and Marble when it surfaced in early December. But by then, I already had a sense something was coming. The first conversation — As someone deep into volumetric capture and radiance fields, I obviously knew about Ben Mildenhall and his pioneering work on NeRF. To my surprise, Ben reached out to me in late June 2024. He’d been following some of my experiments and wanted to chat about my process and workflows and how I was using this “stuff” creatively. At that point he didn’t share what he was building, but we had a genuinely great conversation about radiance fields, AI, and my work. He was curious about the creative perspective, not just the technical one. When the World Labs announcement dropped a few months later, it all made sense. I understood what Ben had been working on, and why the creative angle mattered to them. Then in August 2025, he invited me to try the Marble beta, and I’ve been experimenting with it since. Experimenting with Marble — The first thing I used Marble for was materializing scene and world concepts during ideation at the studio, and seeing if and how it could fit into our production pipeline. In parallel, I dove into a series of experiments focused on world manipulation: starting from real captured spaces and transforming them using Marble. I’d already been exploring that idea using img2img diffusion with ControlNet on NeRF renders, real-time video streams, and even mixed reality using headset camera feeds. But Marble brings something different. It generates persistent, spatially cohesive 3D worlds that can be rendered in real time across a wide range of devices. That’s a real shift. Experiment 01: Parallel Realities — The first experiment, Parallel Realities, starts from a volumetric capture of a real location, reconstructed as 3D Gaussian Splats. Using Marble, I generate an alternate version of that same space, something informed by the original architecture: abandoned, nature-reclaimed, alternate era. Then, using Spark (World Labs’ 3D Gaussian Splatting renderer for THREE.js) I make both realities coexist in the same spatial coordinate system. From there, I use a portal UX mechanic to let the user step between the real reconstruction and the Marble-generated version. Experiment 02: Hidden Depth The second experiment, Hidden Depth, does not transform a space as much as expand it. A captured location has a visual boundary (a mural, a doorway, a dark corridor) and Marble generates what exists beyond it. For example: a Montreal alley has a painted mural; step through it and you’re inside a world informed by what is actually depicted there. World Labs showcased part of this work here: And in their Spark 2.0 post: The project page is here: Why this matters to me — Being able to start from a real 3D Gaussian Splat scene and manipulate it with Marble opens up a lot of ideas. The 3DGS pipeline is becoming an increasingly compelling foundation for exploration, experimentation, and storytelling. What matters most to me right now is more control. The more I can steer the generated scene or world, the more useful the tool becomes. I want more features like the already existing multiple input images and Chisel, the blockout-based approach. I would like better local control, the ability to expand a generated world more and more while preserving coherence, and the ability to directly import 3D Gaussian Splat scenes to be used as a starting point. I want more ways to shape the result, not just a “prompt and hope” approach. — It is exciting to see this field moving from research and demos toward actual creative workflows.

Hugues Bruyère

69,960 просмотров • 1 месяц назад

Over the past two years, AI video models have been competing on realism, resolution, and duration. But no matter how impressive the results look, we remain passive viewers: we press play, watch the clip, and it ends. AlayaWorld Alaya Lab is attempting something fundamentally different. Instead of generating a fixed video, it generates a world that continues to unfold as you move through it. These three demos show the same journey toward a green village rendered in three distinct styles: photorealistic, oil painting, and line art. As the camera moves forward, the model continues generating the road, fences, trees, and distant village. This is not simply an existing video with different filters applied. The environment is generated continuously along the camera trajectory, allowing the scene to develop as the user explores it. AlayaWorld streams video at 720p and 24 FPS while supporting camera movement and viewpoint control. The real breakthrough is not just image quality. Once generation becomes fast enough to respond within an interactive loop, the user is no longer merely watching a video. They become a participant inside the generated world. The world can also respond to new instructions. During generation, users can introduce prompts that trigger spells, summon characters, create explosions, or transform the environment. Most video models follow an initial prompt and produce a predetermined clip. AlayaWorld can respond to changing intent while the world is still running, allowing subsequent events to evolve according to the user’s commands. Generating an attractive frame is relatively easy. Maintaining a coherent world over time is much harder. As a video model repeatedly predicts the next frame, small errors can accumulate until roads, buildings, and objects begin to distort or disappear. AlayaWorld combines spatial memory with compressed historical context, helping the model remember both where things are and what has already happened. This enables stable generation lasting more than one minute while improving consistency when the camera leaves an area and later returns. This may be the next step for AI video: not simply generating a longer movie, but generating a world that can be explored, changed, and interacted with. AlayaWorld is developed by Alaya Lab. The team is progressively releasing its inference code, training code, and datasets, with an online experience expected to launch near the end of the month. Project page:

Over the past two years, AI video models have been competing on realism, resolution, and duration. But no matter how impressive the results look, we remain passive viewers: we press play, watch the clip, and it ends. AlayaWorld Alaya Lab is attempting something fundamentally different. Instead of generating a fixed video, it generates a world that continues to unfold as you move through it. These three demos show the same journey toward a green village rendered in three distinct styles: photorealistic, oil painting, and line art. As the camera moves forward, the model continues generating the road, fences, trees, and distant village. This is not simply an existing video with different filters applied. The environment is generated continuously along the camera trajectory, allowing the scene to develop as the user explores it. AlayaWorld streams video at 720p and 24 FPS while supporting camera movement and viewpoint control. The real breakthrough is not just image quality. Once generation becomes fast enough to respond within an interactive loop, the user is no longer merely watching a video. They become a participant inside the generated world. The world can also respond to new instructions. During generation, users can introduce prompts that trigger spells, summon characters, create explosions, or transform the environment. Most video models follow an initial prompt and produce a predetermined clip. AlayaWorld can respond to changing intent while the world is still running, allowing subsequent events to evolve according to the user’s commands. Generating an attractive frame is relatively easy. Maintaining a coherent world over time is much harder. As a video model repeatedly predicts the next frame, small errors can accumulate until roads, buildings, and objects begin to distort or disappear. AlayaWorld combines spatial memory with compressed historical context, helping the model remember both where things are and what has already happened. This enables stable generation lasting more than one minute while improving consistency when the camera leaves an area and later returns. This may be the next step for AI video: not simply generating a longer movie, but generating a world that can be explored, changed, and interacted with. AlayaWorld is developed by Alaya Lab. The team is progressively releasing its inference code, training code, and datasets, with an online experience expected to launch near the end of the month. Project page:

Rachel🥥

76,960 просмотров • 15 дней назад

Most AI world models impress you for 20 seconds. The harder problem is making them stay coherent. That's what stood out to me in Robbyant's LingBot-World-Infinity. Instead of chasing another resolution headline, the paper focuses on reducing autoregressive drift the issue where small prediction errors compound until the world slowly falls apart. Their stress test is a single uninterrupted 60-minute rollout across 20 different scenarios with no visible quality decay. If you want to see the interactive system yourself, it's live on Reactor: 🎮 I also like that they don't pretend it's perfect. The paper openly says it doesn't have true long-term memory. Leave an area and come back later, and it regenerates something plausible instead of recalling the exact same place. That's a refreshingly honest limitation. Open weights make it even more interesting to dig into. Robbyant For anyone who wants to explore the research: 💻 GitHub: 🤗 Model Weights:

Most AI world models impress you for 20 seconds. The harder problem is making them stay coherent. That's what stood out to me in Robbyant's LingBot-World-Infinity. Instead of chasing another resolution headline, the paper focuses on reducing autoregressive drift the issue where small prediction errors compound until the world slowly falls apart. Their stress test is a single uninterrupted 60-minute rollout across 20 different scenarios with no visible quality decay. If you want to see the interactive system yourself, it's live on Reactor: 🎮 I also like that they don't pretend it's perfect. The paper openly says it doesn't have true long-term memory. Leave an area and come back later, and it regenerates something plausible instead of recalling the exact same place. That's a refreshingly honest limitation. Open weights make it even more interesting to dig into. Robbyant For anyone who wants to explore the research: 💻 GitHub: 🤗 Model Weights:

Shruti Codes

15,066 просмотров • 20 дней назад

Marc Andreessen: “I’m always urging founders to raise prices, raise prices, raise prices.” “We spend a lot of time working with our companies on pricing,” a16z co-founder Marc Andreessen explains. “It’s really this magical art and science that a lot of companies don’t take seriously enough.” Marc continues: “A core principle of pricing is that you don’t want to price by cost if you can avoid it. You want to price by value. Especially when you’re selling to businesses, you want to price as a percentage of the business value you’re creating.” He gives the example of building an AI that can do the job of a programmer, a lawyer, or a radiologist: “Can you price by value and get a percentage of what otherwise would’ve literally been a person? Or equivalently can you price by marginal productivity? If you can take a human doctor and make them much more productive because you give them AI, can you price as a percentage of the productivity uplift?” Marc argues that high prices are under-appreciated by founders: “The naive view on pricing is the lower the pricing, the better it is for the customer. The more sophisticated way of looking at it is that higher prices are often good for the customer because the higher price means the vendor can make the product better, faster. Companies with higher prices and higher margins can actually invest more in R&D and make the product better. Most people who buy things aren’t just looking for the cheapest price. They want something that’s going to work really well.” Marc also emphasizes this point in an interview in Elad Gil’s High Growth Handbook: “What I hear from companies is, ‘Oh, we have an awesome moat, and we’re still going to price our product cheap, because we think that’s somehow going to maximize our business.’ I’m always urging founders to raise prices, raise prices, raise prices. I’m always urging founders to raise prices, raise prices, raise prices. First of all, raising prices is a great way to flesh out whether you actually do have a moat. If you do have a moat, the customers will still buy, because they have to. The definition of a moat is the ability to charge more. And so number one, it’s just a good way to flesh out that topic and really expose it to sunlight. And then number two, companies that charge more can better fund both their distribution efforts and their ongoing R&D efforts. Charging more is a key lever to be able to grow. And the companies that charge more therefore tend to grow faster. That’s counterintuitive to a lot of engineers. A lot of engineers think there’s a one-dimensional relationship between price and value. They have this mental model of commerce like they’re selling rice or something. It’s like, “My product is magical and nobody can replicate it, and I need to price it like it’s a commodity.” No, you don’t. In fact, quite the opposite. If you price it high, then you can fund a much more expensive sales and marketing effort, which means you’re much more likely to win the market, which means you’re much more likely to be able afford to do all the R&D and acquisitions you’re going to want to do. And so we always try to snap people into a two-dimensional mindset, where higher prices equals faster growth.” Video source: a16z (2026)

Marc Andreessen: “I’m always urging founders to raise prices, raise prices, raise prices.” “We spend a lot of time working with our companies on pricing,” a16z co-founder Marc Andreessen explains. “It’s really this magical art and science that a lot of companies don’t take seriously enough.” Marc continues: “A core principle of pricing is that you don’t want to price by cost if you can avoid it. You want to price by value. Especially when you’re selling to businesses, you want to price as a percentage of the business value you’re creating.” He gives the example of building an AI that can do the job of a programmer, a lawyer, or a radiologist: “Can you price by value and get a percentage of what otherwise would’ve literally been a person? Or equivalently can you price by marginal productivity? If you can take a human doctor and make them much more productive because you give them AI, can you price as a percentage of the productivity uplift?” Marc argues that high prices are under-appreciated by founders: “The naive view on pricing is the lower the pricing, the better it is for the customer. The more sophisticated way of looking at it is that higher prices are often good for the customer because the higher price means the vendor can make the product better, faster. Companies with higher prices and higher margins can actually invest more in R&D and make the product better. Most people who buy things aren’t just looking for the cheapest price. They want something that’s going to work really well.” Marc also emphasizes this point in an interview in Elad Gil’s High Growth Handbook: “What I hear from companies is, ‘Oh, we have an awesome moat, and we’re still going to price our product cheap, because we think that’s somehow going to maximize our business.’ I’m always urging founders to raise prices, raise prices, raise prices. I’m always urging founders to raise prices, raise prices, raise prices. First of all, raising prices is a great way to flesh out whether you actually do have a moat. If you do have a moat, the customers will still buy, because they have to. The definition of a moat is the ability to charge more. And so number one, it’s just a good way to flesh out that topic and really expose it to sunlight. And then number two, companies that charge more can better fund both their distribution efforts and their ongoing R&D efforts. Charging more is a key lever to be able to grow. And the companies that charge more therefore tend to grow faster. That’s counterintuitive to a lot of engineers. A lot of engineers think there’s a one-dimensional relationship between price and value. They have this mental model of commerce like they’re selling rice or something. It’s like, “My product is magical and nobody can replicate it, and I need to price it like it’s a commodity.” No, you don’t. In fact, quite the opposite. If you price it high, then you can fund a much more expensive sales and marketing effort, which means you’re much more likely to win the market, which means you’re much more likely to be able afford to do all the R&D and acquisitions you’re going to want to do. And so we always try to snap people into a two-dimensional mindset, where higher prices equals faster growth.” Video source: a16z (2026)

Startup Archive

422,892 просмотров • 6 месяцев назад

WE ARE WATCHING THE TRADITIONAL 3D RENDERING PIPELINE GET REPLACED IN REAL TIME PixVerse just dropped an amazing early experiment called PixVerse Game. Instead of hardcoding mechanics to specific character models, they figured out how to completely decouple a game's underlying math from what you actually see on screen. Here is how their new 3-layer architecture works: ✦ The Engine → handles the raw rules, objectives, and health points. ✦ The Agent → uses AI to translate a player's natural language into valid game moves. ✦ The Visuals → an underlying R1 model generates the world and audio on the fly. Why does this matter? Because you can take the exact same game rules and swap the world instantly. (e.g., swapping from a sci-fi mission to an anime adventure without rewriting the game). This is a fascinating look at where gaming is heading! You can try PixVerse Game here → Currently invite-only, but I snagged a few access codes for y'all :) Code 1: CTQLMY Code 2: LQCAYS Enjoy!

WE ARE WATCHING THE TRADITIONAL 3D RENDERING PIPELINE GET REPLACED IN REAL TIME PixVerse just dropped an amazing early experiment called PixVerse Game. Instead of hardcoding mechanics to specific character models, they figured out how to completely decouple a game's underlying math from what you actually see on screen. Here is how their new 3-layer architecture works: ✦ The Engine → handles the raw rules, objectives, and health points. ✦ The Agent → uses AI to translate a player's natural language into valid game moves. ✦ The Visuals → an underlying R1 model generates the world and audio on the fly. Why does this matter? Because you can take the exact same game rules and swap the world instantly. (e.g., swapping from a sci-fi mission to an anime adventure without rewriting the game). This is a fascinating look at where gaming is heading! You can try PixVerse Game here → Currently invite-only, but I snagged a few access codes for y'all :) Code 1: CTQLMY Code 2: LQCAYS Enjoy!

Charly Wargnier

73,152 просмотров • 16 дней назад

Give your voice agents a face. Introducing Lemon Slice-2 – the world's first interactive talking AI video model Think of it as a face layer for your voice agents, available today in production as an API or an embeddable widget. What makes Lemon Slice-2 great: → Easy: All you need is 1 photo → Expressive: hands and full body motion → Any character and style: if it has a face, Lemon Slice-2 can animate it Under the hood: we trained a custom, 20B-parameter video diffusion transformer, streaming at 20fps on a single GPU. Infinite-length video generation with no error accumulation. Talk to the demo avatars on our homepage. Or, add your own interactive avatar to your website. Companies and creators are already building video agents that are: → Customer support agents who screen-share and show you where to click → Lead qualification agents that replace forms with real conversations → AI personal shoppers who try on clothes and guide purchases People don't engage with chat bubbles. They engage with faces. And now you can have a 24/7 spokesperson that’s the face of your brand. We’ve partnered with , Daily, and Modal to make this happen. As part of this launch, we’re giving away a guide: “How to Add a High-Converting Video Agent to Your Website in under 10 Minutes” Retweet and comment "LEMON SLICE" below and I'll send it over 🍋

Lina Colucci

1,162,861 просмотров • 7 месяцев назад