Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

I usually skip benchmark comparisons. They're almost always picked to flatter. But EgoSchema is different. It measures first-person, egocentric video understanding. The AI equivalent of “look through the robot's eyes and tell me what happened.” I spent some time testing Perceptron Mk1 (“Mark One”) on the robotics workflows Perceptron...

JayminSOfficial's profile picture

Jaymin Shah

87,897 subscribers

116,425 Aufrufe • vor 2 Monaten •via X (Twitter)

Nachrichten & PolitikWissenschaft & Technologie
LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

This looks like another step toward removing the data bottleneck in robotics. The bottleneck exists because recording demonstrations is easy, but labeling every grasp, hold, move, and release is slow and expensive. Perceptron Egocentric dropped a new robotics video annotation system. You give it raw robot-camera or first-person/egocentric video. It gives back structured labels that look more like training data: where each small manipulation starts and ends, what the subtask is, what each hand is doing, where the hands are frame by frame, and which hand is left or right. So, Input: raw video of a person or robot doing a task. Output: a machine-readable breakdown of the physical actions inside that video. Basically, it is a machine that converts robot video into policy-training supervision. That means timestamps, subtask boundaries, per-hand actions, and left-right hand grounding. So this will turn large piles of robot or egocentric video into supervision for training robot policies.
0:58

This looks like another step toward removing the data bottleneck in robotics. The bottleneck exists because recording demonstrations is easy, but labeling every grasp, hold, move, and release is slow and expensive. Perceptron Egocentric dropped a new robotics video annotation system. You give it raw robot-camera or first-person/egocentric video. It gives back structured labels that look more like training data: where each small manipulation starts and ends, what the subtask is, what each hand is doing, where the hands are frame by frame, and which hand is left or right. So, Input: raw video of a person or robot doing a task. Output: a machine-readable breakdown of the physical actions inside that video. Basically, it is a machine that converts robot video into policy-training supervision. That means timestamps, subtask boundaries, per-hand actions, and left-right hand grounding. So this will turn large piles of robot or egocentric video into supervision for training robot policies.

Rohan Paul

18,903 Aufrufe • vor 28 Tagen

Jensen Huang says agentic AI is not a chatbot. It is a "digital robot" -- and the next bottleneck is not the model. It is first-person data. "Agentic AI is just a digital robot. It understands, it reasons, it plans, and it acts and uses tools." Then he makes the bigger jump: "The world's video data is from a third person, not first person. And so agentic systems, robotic systems, physical AI -- the data is the hardest problem." That is the hidden constraint in robotics. LLMs got the internet. Physical AI needs demonstrations, synthetic worlds, reinforcement learning, and first-person experience. This is why Jensen keeps calling it a stack: robots are not waiting for one magical model. They need the data engine, the simulator, the reward system, the runtime, and the computer. The next wave of AI is not just software that talks. It is software that sees, plans, acts, and eventually moves through the world.
2:32

Jensen Huang says agentic AI is not a chatbot. It is a "digital robot" -- and the next bottleneck is not the model. It is first-person data. "Agentic AI is just a digital robot. It understands, it reasons, it plans, and it acts and uses tools." Then he makes the bigger jump: "The world's video data is from a third person, not first person. And so agentic systems, robotic systems, physical AI -- the data is the hardest problem." That is the hidden constraint in robotics. LLMs got the internet. Physical AI needs demonstrations, synthetic worlds, reinforcement learning, and first-person experience. This is why Jensen keeps calling it a stack: robots are not waiting for one magical model. They need the data engine, the simulator, the reward system, the runtime, and the computer. The next wave of AI is not just software that talks. It is software that sees, plans, acts, and eventually moves through the world.

Karl Mehta

14,140 Aufrufe • vor 1 Monat

On Bloomberg last night, I was asked why Amazon drives more than a third of U.S. e-commerce. It is not because of their storefront. It is not because of their payments. It is not even because of product selection. It is because consumers trust their deliveries. That trust is built on physical infrastructure at scale with intelligence layered on top. And that is what Stord is building for every independent merchant. The most defensible companies in the world combine software intelligence with physical execution. Today, 9 of the 10 largest U.S. companies by revenue operate physical networks at scale. Software alone cannot replicate the outcomes created when you control the movement of physical atoms. That is the physical intelligence layer for commerce. And that is what Stord Labs is built to advance. Deploying AI and robotics against real operational complexity, in the real world, across a vertically integrated network operating at scale. Bloomberg TV
5:51

On Bloomberg last night, I was asked why Amazon drives more than a third of U.S. e-commerce. It is not because of their storefront. It is not because of their payments. It is not even because of product selection. It is because consumers trust their deliveries. That trust is built on physical infrastructure at scale with intelligence layered on top. And that is what Stord is building for every independent merchant. The most defensible companies in the world combine software intelligence with physical execution. Today, 9 of the 10 largest U.S. companies by revenue operate physical networks at scale. Software alone cannot replicate the outcomes created when you control the movement of physical atoms. That is the physical intelligence layer for commerce. And that is what Stord Labs is built to advance. Deploying AI and robotics against real operational complexity, in the real world, across a vertically integrated network operating at scale. Bloomberg TV

Sean Henry

1,275,331 Aufrufe • vor 2 Monaten

I am stocked to announce that I won the OpenAI Developers Codex x Mollie Hacka Worldwide Hackathon in Paris. 60+ builders, every one of us working solo, one day to ship. I built mine around a single question: who gets to own intelligence? The default answer is scary. You hand your data to a handful of labs, they train the model, they own it, and you rent back a thin slice of what your own data made possible. That is the bargain on the table today. I do not accept it. So I built Lensemble: a Tapestry like distributed training platform for JEPA based World Models. What does it enable: World Models that a community improves together, keeps sovereign, and co-owns. Two bets sit underneath it. First, the paradigm. Language models predict the next token. Powerful for text, a dead end for the physical world. A robot does not need to autocomplete sentences, it needs to predict what happens next in the world. That is what JEPA does: it learns by predicting representations instead of pixels or tokens. I am convinced world models are the most underrated paradigm in AI right now, and the closest thing we have to a ChatGPT moment for robotics. Second, the politics. Your raw trajectories never leave your machine. Each participant trains locally against a shared protocol and ships only an update, never the data. A federated round folds those updates into one shared world model, a LeWorldModel based model, and the gain is measured, not claimed: a 12k-parameter adapter on a frozen backbone, held-out prediction error down about 12 percent, the model measurably less surprised by the world. Then the upside is split by contribution weight, so the people who improved the model own a share of what it earns. This is the thesis behind Project Tapestry, the AI Alliance and Yann LeCun's push for federated, sovereign frontier AI, carried into world models and robotics. Call it Tapestry for the physical world. All of it built solo, in a single day, with Codex as my pair the whole way. Thank you to OpenAI Codex and Mollie for backing builders who ship real things, and to Boris and the organizing crew for the room and the standard you set. Intelligence the world improves, and the world owns. That is the future I want for my kids, and the one I will keep building.
1:13

I am stocked to announce that I won the OpenAI Developers Codex x Mollie Hacka Worldwide Hackathon in Paris. 60+ builders, every one of us working solo, one day to ship. I built mine around a single question: who gets to own intelligence? The default answer is scary. You hand your data to a handful of labs, they train the model, they own it, and you rent back a thin slice of what your own data made possible. That is the bargain on the table today. I do not accept it. So I built Lensemble: a Tapestry like distributed training platform for JEPA based World Models. What does it enable: World Models that a community improves together, keeps sovereign, and co-owns. Two bets sit underneath it. First, the paradigm. Language models predict the next token. Powerful for text, a dead end for the physical world. A robot does not need to autocomplete sentences, it needs to predict what happens next in the world. That is what JEPA does: it learns by predicting representations instead of pixels or tokens. I am convinced world models are the most underrated paradigm in AI right now, and the closest thing we have to a ChatGPT moment for robotics. Second, the politics. Your raw trajectories never leave your machine. Each participant trains locally against a shared protocol and ships only an update, never the data. A federated round folds those updates into one shared world model, a LeWorldModel based model, and the gain is measured, not claimed: a 12k-parameter adapter on a frozen backbone, held-out prediction error down about 12 percent, the model measurably less surprised by the world. Then the upside is split by contribution weight, so the people who improved the model own a share of what it earns. This is the thesis behind Project Tapestry, the AI Alliance and Yann LeCun's push for federated, sovereign frontier AI, carried into world models and robotics. Call it Tapestry for the physical world. All of it built solo, in a single day, with Codex as my pair the whole way. Thank you to OpenAI Codex and Mollie for backing builders who ship real things, and to Boris and the organizing crew for the room and the standard you set. Intelligence the world improves, and the world owns. That is the future I want for my kids, and the one I will keep building.

abdel

17,370 Aufrufe • vor 1 Monat

SITUATION EXPLAINED: A new generation of AI labs are betting that world models are the path to general intelligence that can interact with the real world. We had General Intuition CEO Pim de Witte on MTS to explain what world models are, and why intuitively world models offer a clearer path towards general intelligence than LLMs. For humans, text is a technology for compressing information about the unfolding, three-dimensional world around us and communicating it to other people. When LLMs reason about the world, they do so by reconstructing the world through text. World models work by modeling the three-dimensional world at the architecture level and then allowing agents to interact with that world. Unlike an LLM, which understands space and time through text, world models perform spatial and temporal reasoning natively. General Intuition hopes to use world models to create embodied AI—machines that can reason and act in the real world. But other companies are also pushing the frontier in this space. Odyssey ML today announced a multi-agent world model, Agora-1. Their launch video shows four participants interacting with a real-time world model simulation of GoldenEye, a first-person shooter for the N64. A playable preview of the GoldenEye simulation is available on their site.
1:31

SITUATION EXPLAINED: A new generation of AI labs are betting that world models are the path to general intelligence that can interact with the real world. We had General Intuition CEO Pim de Witte on MTS to explain what world models are, and why intuitively world models offer a clearer path towards general intelligence than LLMs. For humans, text is a technology for compressing information about the unfolding, three-dimensional world around us and communicating it to other people. When LLMs reason about the world, they do so by reconstructing the world through text. World models work by modeling the three-dimensional world at the architecture level and then allowing agents to interact with that world. Unlike an LLM, which understands space and time through text, world models perform spatial and temporal reasoning natively. General Intuition hopes to use world models to create embodied AI—machines that can reason and act in the real world. But other companies are also pushing the frontier in this space. Odyssey ML today announced a multi-agent world model, Agora-1. Their launch video shows four participants interacting with a real-time world model simulation of GoldenEye, a first-person shooter for the N64. A playable preview of the GoldenEye simulation is available on their site.

MTS

11,641 Aufrufe • vor 2 Monaten

Over the past year, a lot of the conversation has been around how quickly AI is changing the digital world. At the same time, it made us think about how little has changed in the physical one. Most products are still designed the same way, just with more features layered on top. We wanted to explore a different direction at Lusion. Instead of asking how to add intelligence, we asked what happens if you take things away. What remains when you strip a product back to its core purpose, and design it properly from there. Oryzo is our first attempt at that. It is a physical product, built with a very simple idea of how it should exist and behave. No dependency on external systems, no complexity for the sake of it. Just something that feels considered. Curious to hear what people think. Link is in the comment!
1:02

Over the past year, a lot of the conversation has been around how quickly AI is changing the digital world. At the same time, it made us think about how little has changed in the physical one. Most products are still designed the same way, just with more features layered on top. We wanted to explore a different direction at Lusion. Instead of asking how to add intelligence, we asked what happens if you take things away. What remains when you strip a product back to its core purpose, and design it properly from there. Oryzo is our first attempt at that. It is a physical product, built with a very simple idea of how it should exist and behave. No dependency on external systems, no complexity for the sake of it. Just something that feels considered. Curious to hear what people think. Link is in the comment!

Lusion™

86,926 Aufrufe • vor 4 Monaten

🚀 Three Next-Gen AI & Web3 Projects Are Launching on Mindo AI A new chapter for community-powered intelligence, prediction markets, and open AI infrastructure The AI + Web3 landscape is entering a decisive phase — one where real usage, real revenue, and real ownership matter more than hype. Today, MindoAI is proud to welcome three groundbreaking projects that represent this shift clearly and powerfully: Perceptron Network Space DeepNode AI Each project tackles a different bottleneck in the AI economy — data, forecasting, and infrastructure — but they all share the same vision: decentralization, community ownership, and sustainable value creation. Let’s take a deeper look 👇 🧠 Perceptron Network The world’s first community-powered AI data engine Perceptron Network is redefining how AI data is sourced, validated, and delivered. Instead of relying on expensive, closed, and slow legacy data providers, Perceptron unlocks community-powered data pipelines that are: Faster Cheaper Revenue-generating from day one This isn’t experimental AI infrastructure — Perceptron already serves real clients with real revenue, proving that decentralized data engines can outperform traditional incumbents. Why Perceptron matters: AI models are only as good as their data Centralized data monopolies slow innovation Communities can produce higher-quality data at scale By aligning contributors, validators, and clients through incentives, Perceptron turns unused human and network potential into a living data engine for AI. Launching on Mindo AI gives Perceptron access to a broader AI-native community — accelerating adoption, partnerships, and ecosystem growth. 🌌 intodotspace The first 10× leveraged prediction market on Solana intodotspace is pushing the boundaries of on-chain prediction markets. Built by the $1.5B UFO team, this platform introduces: 10× leveraged predictions Ultra-fast execution on Solana Deep liquidity and composable market design The market’s confidence is already clear — the project completed a record-breaking raise that was oversubscribed by 1,360%. What makes intodotspace different: Leverage amplifies conviction, not noise On-chain transparency replaces opaque odds Markets become real-time intelligence engines Prediction markets are often called “truth machines.” intodotspace upgrades them into high-signal, high-efficiency forecasting layers — useful for traders, protocols, DAOs, and even AI systems that need probabilistic insights. Launching on positions intodotspace at the intersection of AI-driven decision-making and on-chain market intelligence. 🌐 DeepNode AI Infrastructure for open intelligence DeepNode AI is tackling one of the biggest problems in modern AI: centralized ownership. Today, AI is dominated by a handful of corporations. DeepNode flips that model by building open intelligence infrastructure where: Anyone can deploy AI models Builders earn directly from usage Intelligence is co-owned, not extracted Backed by leading validators, miners, and ecosystem builders, DeepNode transforms AI from a closed monopoly into a shared utility. DeepNode’s core philosophy: “Own what you build — or someone else will.” This is more than infrastructure. It’s an economic redesign of AI itself: Builders keep ownership Contributors share upside Networks replace platforms Launching on connects DeepNode to creators, researchers, and communities who believe intelligence should belong to everyone — not just Big Tech. 🤝 Why This Matters for With the launch of Perceptron Network, intodotspace, and DeepNode AI, #MindoAI is rapidly becoming: A hub for AI-native Web3 innovation A launchpad for real, revenue-backed projects A meeting point for data, markets, and intelligence infrastructure These three projects don’t compete — they complement each other: Perceptron supplies data intodotspace produces market intelligence DeepNode powers open AI execution Together, they form the backbone of a decentralized intelligence economy. 🔥 The future of AI is open, composable, and community-owned — and it’s launching now on Which of these projects are you most excited about? And how do you see decentralized intelligence reshaping the next AI cycle? 👇 Share your thoughts and join the conversation.
0:38

🚀 Three Next-Gen AI & Web3 Projects Are Launching on Mindo AI A new chapter for community-powered intelligence, prediction markets, and open AI infrastructure The AI + Web3 landscape is entering a decisive phase — one where real usage, real revenue, and real ownership matter more than hype. Today, MindoAI is proud to welcome three groundbreaking projects that represent this shift clearly and powerfully: Perceptron Network Space DeepNode AI Each project tackles a different bottleneck in the AI economy — data, forecasting, and infrastructure — but they all share the same vision: decentralization, community ownership, and sustainable value creation. Let’s take a deeper look 👇 🧠 Perceptron Network The world’s first community-powered AI data engine Perceptron Network is redefining how AI data is sourced, validated, and delivered. Instead of relying on expensive, closed, and slow legacy data providers, Perceptron unlocks community-powered data pipelines that are: Faster Cheaper Revenue-generating from day one This isn’t experimental AI infrastructure — Perceptron already serves real clients with real revenue, proving that decentralized data engines can outperform traditional incumbents. Why Perceptron matters: AI models are only as good as their data Centralized data monopolies slow innovation Communities can produce higher-quality data at scale By aligning contributors, validators, and clients through incentives, Perceptron turns unused human and network potential into a living data engine for AI. Launching on Mindo AI gives Perceptron access to a broader AI-native community — accelerating adoption, partnerships, and ecosystem growth. 🌌 intodotspace The first 10× leveraged prediction market on Solana intodotspace is pushing the boundaries of on-chain prediction markets. Built by the $1.5B UFO team, this platform introduces: 10× leveraged predictions Ultra-fast execution on Solana Deep liquidity and composable market design The market’s confidence is already clear — the project completed a record-breaking raise that was oversubscribed by 1,360%. What makes intodotspace different: Leverage amplifies conviction, not noise On-chain transparency replaces opaque odds Markets become real-time intelligence engines Prediction markets are often called “truth machines.” intodotspace upgrades them into high-signal, high-efficiency forecasting layers — useful for traders, protocols, DAOs, and even AI systems that need probabilistic insights. Launching on positions intodotspace at the intersection of AI-driven decision-making and on-chain market intelligence. 🌐 DeepNode AI Infrastructure for open intelligence DeepNode AI is tackling one of the biggest problems in modern AI: centralized ownership. Today, AI is dominated by a handful of corporations. DeepNode flips that model by building open intelligence infrastructure where: Anyone can deploy AI models Builders earn directly from usage Intelligence is co-owned, not extracted Backed by leading validators, miners, and ecosystem builders, DeepNode transforms AI from a closed monopoly into a shared utility. DeepNode’s core philosophy: “Own what you build — or someone else will.” This is more than infrastructure. It’s an economic redesign of AI itself: Builders keep ownership Contributors share upside Networks replace platforms Launching on connects DeepNode to creators, researchers, and communities who believe intelligence should belong to everyone — not just Big Tech. 🤝 Why This Matters for With the launch of Perceptron Network, intodotspace, and DeepNode AI, #MindoAI is rapidly becoming: A hub for AI-native Web3 innovation A launchpad for real, revenue-backed projects A meeting point for data, markets, and intelligence infrastructure These three projects don’t compete — they complement each other: Perceptron supplies data intodotspace produces market intelligence DeepNode powers open AI execution Together, they form the backbone of a decentralized intelligence economy. 🔥 The future of AI is open, composable, and community-owned — and it’s launching now on Which of these projects are you most excited about? And how do you see decentralized intelligence reshaping the next AI cycle? 👇 Share your thoughts and join the conversation.

Hồng Ngọc | Ruby💎

12,837 Aufrufe • vor 5 Monaten

I Tried 6 AI Video Editors. NemoVideo Is the Only One I Actually Kept Using. I've been creating short-form content for almost two years. In that time, I've tested CapCut, OpenClaw, and a handful of other tools that claimed to be the "best AI video editor" on the market. Most of them were either too complicated, too expensive, or just… not built for real creators. NemoVideo is different. It doesn't try to do everything. It focuses on one thing: taking your raw footage and turning it into a polished, publish-ready video — fast. The AI rough cut feature alone saves me 60%+ of my editing time. I upload my clips, the AI finds the best moments, trims the dead air, and hands me something I can actually work with. No manual scrubbing through hours of footage. The auto-caption tool is just as good. Accurate, fast, and it exports cleanly without watermarks — even on the free plan. If you're a creator who's tired of spending more time editing than actually creating, give NemoVideo a shot. →
0:50

I Tried 6 AI Video Editors. NemoVideo Is the Only One I Actually Kept Using. I've been creating short-form content for almost two years. In that time, I've tested CapCut, OpenClaw, and a handful of other tools that claimed to be the "best AI video editor" on the market. Most of them were either too complicated, too expensive, or just… not built for real creators. NemoVideo is different. It doesn't try to do everything. It focuses on one thing: taking your raw footage and turning it into a polished, publish-ready video — fast. The AI rough cut feature alone saves me 60%+ of my editing time. I upload my clips, the AI finds the best moments, trims the dead air, and hands me something I can actually work with. No manual scrubbing through hours of footage. The auto-caption tool is just as good. Accurate, fast, and it exports cleanly without watermarks — even on the free plan. If you're a creator who's tired of spending more time editing than actually creating, give NemoVideo a shot. →

NAYAK

30,562 Aufrufe • vor 3 Monaten

Dr. Fei-Fei Li just called out the biggest blind spot in the entire AI industry. We have been building half of human intelligence. And calling it the finish line. Li: “If you look at human intelligence, it pretty much boils down to two buckets.” The first bucket is language. Symbolic reasoning. Communication. The ability to think in words and abstractions. That’s what every major AI lab has spent the last decade building. The second bucket is the one the industry has almost entirely ignored. Li: “We call that in AI spatial intelligence.” How humans and animals perceive, navigate, and interact with the three-dimensional physical world. How we reach for objects. How we move through space. How we build and manipulate physical reality. From painting masterpieces to constructing the pyramids, non-verbal spatial intelligence is what actually shapes the world. Language describes reality. Spatial intelligence acts on it. And the gap between those two things is the gap between a chatbot and a robot. Li: “When this technology is ready, the robotic revolution is gonna start. We’re already seeing that trend.” Every robot is a moving agent. Every moving agent requires spatial intelligence to function in the real world. The humanoid robots being deployed in factories right now are hitting the ceiling of what language models alone can power. Spatial intelligence is the unlock. But Li didn’t stop at robotics. Li: “From a geopolitics point of view, this is part of the technology that goes straight into weapons.” Autonomous drone swarms. Battlefield navigation. Physical target acquisition without human oversight. Every military application of AI that operates in the real world runs on spatial intelligence. The nation that masters the transition from static text to dynamic three-dimensional perception doesn’t just win the software race. It commands the physical battlefield. The AI arms race just broke out of the data center. It’s operating in three dimensions now.
3:08

Dr. Fei-Fei Li just called out the biggest blind spot in the entire AI industry. We have been building half of human intelligence. And calling it the finish line. Li: “If you look at human intelligence, it pretty much boils down to two buckets.” The first bucket is language. Symbolic reasoning. Communication. The ability to think in words and abstractions. That’s what every major AI lab has spent the last decade building. The second bucket is the one the industry has almost entirely ignored. Li: “We call that in AI spatial intelligence.” How humans and animals perceive, navigate, and interact with the three-dimensional physical world. How we reach for objects. How we move through space. How we build and manipulate physical reality. From painting masterpieces to constructing the pyramids, non-verbal spatial intelligence is what actually shapes the world. Language describes reality. Spatial intelligence acts on it. And the gap between those two things is the gap between a chatbot and a robot. Li: “When this technology is ready, the robotic revolution is gonna start. We’re already seeing that trend.” Every robot is a moving agent. Every moving agent requires spatial intelligence to function in the real world. The humanoid robots being deployed in factories right now are hitting the ceiling of what language models alone can power. Spatial intelligence is the unlock. But Li didn’t stop at robotics. Li: “From a geopolitics point of view, this is part of the technology that goes straight into weapons.” Autonomous drone swarms. Battlefield navigation. Physical target acquisition without human oversight. Every military application of AI that operates in the real world runs on spatial intelligence. The nation that masters the transition from static text to dynamic three-dimensional perception doesn’t just win the software race. It commands the physical battlefield. The AI arms race just broke out of the data center. It’s operating in three dimensions now.

Dustin

122,680 Aufrufe • vor 5 Monaten

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

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

Max Zhaoshuo Li 李赵硕

1,078,282 Aufrufe • vor 2 Monaten

MiniMax H3 is native in ComfyUI, day zero with the weights. Model highlights: → Text-to-video, prompt only → Image-to-video → First-and-last-frame - control the opening frame, the closing frame, or both → Reference-to-video - carry a subject, a motion, or a voice through the clip → In-place editing - modify a shot you already have instead of regenerating it The capability MiniMax (official) leads with is multimodal context understanding, and it's what collapses those five tasks into one model. H3 takes images, audio, and video together and resolves them against a prompt that explains how they relate. Download the workflows & learn more in our blog 👇
1:06

MiniMax H3 is native in ComfyUI, day zero with the weights. Model highlights: → Text-to-video, prompt only → Image-to-video → First-and-last-frame - control the opening frame, the closing frame, or both → Reference-to-video - carry a subject, a motion, or a voice through the clip → In-place editing - modify a shot you already have instead of regenerating it The capability MiniMax (official) leads with is multimodal context understanding, and it's what collapses those five tasks into one model. H3 takes images, audio, and video together and resolves them against a prompt that explains how they relate. Download the workflows & learn more in our blog 👇

ComfyUI

216,527 Aufrufe • vor 4 Tagen

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

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

Turing Post

13,836 Aufrufe • vor 2 Monaten

Launching on Wednesday, Sept 25. Stay tuned! Right now, #AI is having a moment — and it’s not the first time grand predictions about the potential of machines are being made. But, what does it really mean to say something like ChatGPT is “intelligent”? What exactly is #intelligence? The next season of the #Complexity #podcast, 'The Nature of Intelligence', explores this question through conversations with cognitive and neuroscientists, animal cognition researchers, and AI experts in six episodes. Together, we'll investigate the complexities of human intelligence, how it compares to that of other species, and where AI fits in. We'll dive into the relationship between language and thought, examine AI's limitations, and ask: Could machines ever truly be like us?
3:00

Launching on Wednesday, Sept 25. Stay tuned! Right now, #AI is having a moment — and it’s not the first time grand predictions about the potential of machines are being made. But, what does it really mean to say something like ChatGPT is “intelligent”? What exactly is #intelligence? The next season of the #Complexity #podcast, 'The Nature of Intelligence', explores this question through conversations with cognitive and neuroscientists, animal cognition researchers, and AI experts in six episodes. Together, we'll investigate the complexities of human intelligence, how it compares to that of other species, and where AI fits in. We'll dive into the relationship between language and thought, examine AI's limitations, and ask: Could machines ever truly be like us?

Santa Fe Institute

45,386 Aufrufe • vor 1 Jahr

AI has transformed how video is created. We think the next wave is about understanding it. Over the past few years, we've seen remarkable advances in video generation, editing, avatars, and creative tooling. An increasingly important problem is teaching machines to search, analyze, reason over, and extract insight from video - across massive libraries and live streams alike. We're calling this video intelligence, and we're actively looking to back founders building here. We're most excited about companies pushing on the core capabilities: - Video-native models - multimodal embeddings, temporal reasoning, and retrieval built specifically for video rather than adapted from image or text - Real-time and large-scale pipelines - infrastructure for processing, indexing, and querying video at the speed and scale enterprises actually need - Agentic and reasoning layers - systems that don't just retrieve clips but answer questions, surface anomalies, and take action on what they see The models and infrastructure to make this real are appearing to be crossing a capability threshold right now. Multimodal foundation models are maturing, storage costs have collapsed, and enterprises are sitting on years of unstructured video with no way to use it. That infrastructure unlocks a wide range of applications including media and sports workflows, security and physical operations, enterprise knowledge management, advertising analytics, robotics, and consumer products, where video has historically been dark data. If you're building in video intelligence at the model layer, the platform layer, or in a vertical application, we'd love to talk!
1:41

AI has transformed how video is created. We think the next wave is about understanding it. Over the past few years, we've seen remarkable advances in video generation, editing, avatars, and creative tooling. An increasingly important problem is teaching machines to search, analyze, reason over, and extract insight from video - across massive libraries and live streams alike. We're calling this video intelligence, and we're actively looking to back founders building here. We're most excited about companies pushing on the core capabilities: - Video-native models - multimodal embeddings, temporal reasoning, and retrieval built specifically for video rather than adapted from image or text - Real-time and large-scale pipelines - infrastructure for processing, indexing, and querying video at the speed and scale enterprises actually need - Agentic and reasoning layers - systems that don't just retrieve clips but answer questions, surface anomalies, and take action on what they see The models and infrastructure to make this real are appearing to be crossing a capability threshold right now. Multimodal foundation models are maturing, storage costs have collapsed, and enterprises are sitting on years of unstructured video with no way to use it. That infrastructure unlocks a wide range of applications including media and sports workflows, security and physical operations, enterprise knowledge management, advertising analytics, robotics, and consumer products, where video has historically been dark data. If you're building in video intelligence at the model layer, the platform layer, or in a vertical application, we'd love to talk!

Jason Cui

36,205 Aufrufe • vor 3 Monaten

most AI APIs either cost too much or lock the best models behind expensive plans. then i came across Agnes AI. it’s one of the few platforms giving developers access to text, image, and video AI models through a single API. their ecosystem gives you direct access to top-tier models like - Agnes-2.0-Flash for text - Agnes-Image-2.0 for image - Agnes-Video-V2.0 for video Instead of managing different providers for different tasks, everything works from one place. what stood out to me: > one API for text, image, and video generation > OpenAI-compatible API, so integration is straightforward > built for speed and scalability > free access for developers to start building > useful for everything from chatbots to image generation and content automation if you’re building AI products, it’s worth taking a look while it’s still flying under the radar. sometimes the biggest advantage is finding great tools before everyone else does. check out Agnes AI here, get free access for developers : and if you’re building with it, join the developer community here: Agnes AI #AgnesAI #AgnesAPI #BuildWithAgnes #MultimodalAI #AIforDevs #DeveloperAPI #AIParity
1:37

most AI APIs either cost too much or lock the best models behind expensive plans. then i came across Agnes AI. it’s one of the few platforms giving developers access to text, image, and video AI models through a single API. their ecosystem gives you direct access to top-tier models like - Agnes-2.0-Flash for text - Agnes-Image-2.0 for image - Agnes-Video-V2.0 for video Instead of managing different providers for different tasks, everything works from one place. what stood out to me: > one API for text, image, and video generation > OpenAI-compatible API, so integration is straightforward > built for speed and scalability > free access for developers to start building > useful for everything from chatbots to image generation and content automation if you’re building AI products, it’s worth taking a look while it’s still flying under the radar. sometimes the biggest advantage is finding great tools before everyone else does. check out Agnes AI here, get free access for developers : and if you’re building with it, join the developer community here: Agnes AI #AgnesAI #AgnesAPI #BuildWithAgnes #MultimodalAI #AIforDevs #DeveloperAPI #AIParity

MIKE

14,591 Aufrufe • vor 17 Tagen

AI video is quietly shifting from generation to direction. After testing Ray 3.2 on a restaurant launch concept, that became pretty obvious. I started with two static images: one exterior shot, one interior dining scene. A few minutes later, I had a cinematic sequence that actually felt connected. The biggest improvement isn't just realism. It's control. Most AI video workflows still look like this: prompt → generate → regenerate → regenerate again. You're mostly hoping the model understands what you want. Ray 3.2 feels different. Instead of generating isolated clips, you can guide how one scene evolves into the next. You can shape transitions, control camera movement, preserve visual consistency, and build sequences that feel intentionally directed. For a simple restaurant commercial, that makes a huge difference. The exterior naturally flows into the dining experience. The pacing feels deliberate. The motion feels cinematic. The story feels connected. And that's what stood out to me most. For the first time, AI video felt less like prompting and more like directing. Every frame, yours to direct. Try Luma today using my link:
0:31

AI video is quietly shifting from generation to direction. After testing Ray 3.2 on a restaurant launch concept, that became pretty obvious. I started with two static images: one exterior shot, one interior dining scene. A few minutes later, I had a cinematic sequence that actually felt connected. The biggest improvement isn't just realism. It's control. Most AI video workflows still look like this: prompt → generate → regenerate → regenerate again. You're mostly hoping the model understands what you want. Ray 3.2 feels different. Instead of generating isolated clips, you can guide how one scene evolves into the next. You can shape transitions, control camera movement, preserve visual consistency, and build sequences that feel intentionally directed. For a simple restaurant commercial, that makes a huge difference. The exterior naturally flows into the dining experience. The pacing feels deliberate. The motion feels cinematic. The story feels connected. And that's what stood out to me most. For the first time, AI video felt less like prompting and more like directing. Every frame, yours to direct. Try Luma today using my link:

SONIA

41,651 Aufrufe • vor 1 Monat

Godfather of AI Yann LeCun reveals the humanoid robot industry's secret: the companies building the robots have no idea how to make them smart enough to be useful. "There is a lot of expectations that robotics is going to be a revolution, essentially, in a big market, eventually." "But the secret of the industry is that none of the companies at the moment that are building robots, like humanoid robots, none of those companies has any idea how to make those robots smart enough to be useful." "There's a big push for AI to make progress towards... understanding the physical world. That's what is prompting this investment." "There's nothing wrong with investing in LLMs. LLMs are useful. They're just addressing a completely different set of applications, really." LeCun is not describing someone else's problem. He has spent 15 years on world models, raised more than $1 billion for AMI Labs to build that missing intelligence layer, and in the same interview he calls 2026 the year of the world model and puts consumer robots 3 to 5 years out. Every humanoid robot demo you have seen is a body waiting for a brain. The winner in robotics will not be the company that shipped the body; it will be the one that ships the intelligence. - Yann LeCun (Yann LeCun), Turing Award winner and founder of AMI Labs, on Bloomberg Television.
1:01

Godfather of AI Yann LeCun reveals the humanoid robot industry's secret: the companies building the robots have no idea how to make them smart enough to be useful. "There is a lot of expectations that robotics is going to be a revolution, essentially, in a big market, eventually." "But the secret of the industry is that none of the companies at the moment that are building robots, like humanoid robots, none of those companies has any idea how to make those robots smart enough to be useful." "There's a big push for AI to make progress towards... understanding the physical world. That's what is prompting this investment." "There's nothing wrong with investing in LLMs. LLMs are useful. They're just addressing a completely different set of applications, really." LeCun is not describing someone else's problem. He has spent 15 years on world models, raised more than $1 billion for AMI Labs to build that missing intelligence layer, and in the same interview he calls 2026 the year of the world model and puts consumer robots 3 to 5 years out. Every humanoid robot demo you have seen is a body waiting for a brain. The winner in robotics will not be the company that shipped the body; it will be the one that ships the intelligence. - Yann LeCun (Yann LeCun), Turing Award winner and founder of AMI Labs, on Bloomberg Television.

Karl Mehta

44,165 Aufrufe • vor 19 Tagen

Chinese robotics company Astribot released their latest World-Action Model (WAM), Lumo-2. Technical breakdown: - based on a frozen 🥶 Qwen-3.5 4B VLM - trained in 3 progressive stages: 1. Action is aligned with latent world dynamics (an abstract representation of action). Real-world actions are anchored to physical constraints, while the latent space is guided to focus on motion-relevant changes. This bidirectional relationship makes the model physically grounded -> critical for a world model. 2. Action is aligned with vision and language. Reusing the vision backbone and action encoder from the frozen VLM, the authors add a custom vocabulary (for new actions), a semantic module, an action decoder, and an action projector. This aligns the (new) action representations with the (existing) vision-language semantic space. Most importantly: it builds a direct mapping from natural-language instructions to motor execution. 3. End-to-end training on language, video, and robot data. Only the new modules (everything outside the frozen backbone) are trained end-to-end across temporal reasoning, physical understanding, long-horizon, and dexterous manipulation. At the end of the day, Lumo-2 is not the best on benchmarks, but that's not the point. What's genuinely new: - a way to combine latent world modeling and action generation through progressive alignment - a physically-grounded latent dynamics space - it lifts performance on unseen objects using un-annotated human egocentric video + Vision Pro captures, no special transfer algorithm needed Why it matters: - the whole model is thin trainable adapters (semantic module, action decoder/projector) on a frozen 4B backbone (cheap) - that scale is suited for real-time embedded inference (~2.71× decode speedup, no accuracy loss) - its real moat is long-horizon execution, where the added temporal memory pays off far more than on any other task As a result, this robot can now make your latte (5x sped up video):
0:39

Chinese robotics company Astribot released their latest World-Action Model (WAM), Lumo-2. Technical breakdown: - based on a frozen 🥶 Qwen-3.5 4B VLM - trained in 3 progressive stages: 1. Action is aligned with latent world dynamics (an abstract representation of action). Real-world actions are anchored to physical constraints, while the latent space is guided to focus on motion-relevant changes. This bidirectional relationship makes the model physically grounded -> critical for a world model. 2. Action is aligned with vision and language. Reusing the vision backbone and action encoder from the frozen VLM, the authors add a custom vocabulary (for new actions), a semantic module, an action decoder, and an action projector. This aligns the (new) action representations with the (existing) vision-language semantic space. Most importantly: it builds a direct mapping from natural-language instructions to motor execution. 3. End-to-end training on language, video, and robot data. Only the new modules (everything outside the frozen backbone) are trained end-to-end across temporal reasoning, physical understanding, long-horizon, and dexterous manipulation. At the end of the day, Lumo-2 is not the best on benchmarks, but that's not the point. What's genuinely new: - a way to combine latent world modeling and action generation through progressive alignment - a physically-grounded latent dynamics space - it lifts performance on unseen objects using un-annotated human egocentric video + Vision Pro captures, no special transfer algorithm needed Why it matters: - the whole model is thin trainable adapters (semantic module, action decoder/projector) on a frozen 4B backbone (cheap) - that scale is suited for real-time embedded inference (~2.71× decode speedup, no accuracy loss) - its real moat is long-horizon execution, where the added temporal memory pays off far more than on any other task As a result, this robot can now make your latte (5x sped up video):

Léo

32,044 Aufrufe • vor 14 Tagen

FRIEDBERG: VIDEO DATA WILL POWER THE NEXT GENERATION OF AI Friedberg broke down the scale shift coming to artificial intelligence, arguing that text based models like GPT are just the beginning, and that the real revolution will come from video-trained systems: “The internet and all these LLMs are language models trained on text from the internet, around 50 billion words total, maybe one to five terabytes of data in their training sets. But if you look at the video data out there, there are hundreds of billions of hours, much of it on YouTube. By some estimates, there’s a thousand exabytes of video data on the internet, about a billion times more than text data. I think we just saw that play out with the new video model that launched yesterday. Google has all this YouTube data, whether or not they’re using it to train, I don’t know. I’ve heard from insiders they’re not allowed to yet and would have to redo the terms of service.” Source: AIFinInsights david friedberg
2:12

FRIEDBERG: VIDEO DATA WILL POWER THE NEXT GENERATION OF AI Friedberg broke down the scale shift coming to artificial intelligence, arguing that text based models like GPT are just the beginning, and that the real revolution will come from video-trained systems: “The internet and all these LLMs are language models trained on text from the internet, around 50 billion words total, maybe one to five terabytes of data in their training sets. But if you look at the video data out there, there are hundreds of billions of hours, much of it on YouTube. By some estimates, there’s a thousand exabytes of video data on the internet, about a billion times more than text data. I think we just saw that play out with the new video model that launched yesterday. Google has all this YouTube data, whether or not they’re using it to train, I don’t know. I’ve heard from insiders they’re not allowed to yet and would have to redo the terms of service.” Source: AIFinInsights david friedberg

Mario Nawfal

15,855 Aufrufe • vor 8 Monaten

Better data for robot hands! 🤲🏼 Embodied reasoning is the bottleneck now. It’s the data we are missing. Specifically, understanding what the robot's hands are doing throughout a task. Perceptron AI new Egocentric offering tracks both hands through an entire video instead of guessing from sampled frames. That matters because a lot happens between frames. Hand positions shift. Contacts change. The robot adjusts grip. You miss all that if you're sampling. 0.280 semantic end-to-end F1 score on WGO-Bench vs 0.158 from the best pipelines built on Gemini Robotics-ER 1.6. Training a robot to manipulate objects requires understanding every moment of hand-object interaction. If your training data has gaps: missing frames, missed contacts, unclear hand positions, the policy learns from incomplete information. You end up with robots that work in demos but fail in real tasks. Continuous hand tracking gives you the full picture. The policy learns from cleaner, more complete data. That compounds into better real-world performance. ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
0:25

Better data for robot hands! 🤲🏼 Embodied reasoning is the bottleneck now. It’s the data we are missing. Specifically, understanding what the robot's hands are doing throughout a task. Perceptron AI new Egocentric offering tracks both hands through an entire video instead of guessing from sampled frames. That matters because a lot happens between frames. Hand positions shift. Contacts change. The robot adjusts grip. You miss all that if you're sampling. 0.280 semantic end-to-end F1 score on WGO-Bench vs 0.158 from the best pipelines built on Gemini Robotics-ER 1.6. Training a robot to manipulate objects requires understanding every moment of hand-object interaction. If your training data has gaps: missing frames, missed contacts, unclear hand positions, the policy learns from incomplete information. You end up with robots that work in demos but fail in real tasks. Continuous hand tracking gives you the full picture. The policy learns from cleaner, more complete data. That compounds into better real-world performance. ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

18,261 Aufrufe • vor 28 Tagen