5 chapters. 17 industry experts. A look at Europe’s AI future. Meet the voices setting the stage at the #NVIDIAGTC Berlin keynote pregame show, exploring sovereign AI, enterprise intelligence, AI safety, AI factories, physical AI, and the growth engines shaping Europe’s future. Watch live October 21 at 9 a.m. CEST → nvda.ws/4i4Zqy2

Hardware Partner Spotlight: NVIDIA joins The @Google Gemma Hardware Hackathon @ AGI House! Build with Gemma. Run it locally on @NVIDIARobotics Jetson. Bring your next edge or physical AI idea to life. On October 24, join 100 selected builders, the @GoogleDeepMind Gemma team, and hardware partners for a full day of hands-on building—from camera assistants and intelligent edge devices to robotics applications. 🤖 NVIDIA hardware available to borrow during the hackathon: • 15 Jetson Orin Nano Super developer kits, each with an NVMe SSD • 15 Jetson Thor developer kits These kits are available for on-site development and must be returned after the event. Availability is limited. 🏆NVIDIA hardware awards are also in the works—stay tuned for the final prize details. Every project must use a Gemma-based model and run entirely on-device. Bring your laptop, your ideas, and your robot if you have one. Compete for a share of $45,000 in cash prizes, plus partner hardware awards. 📅 Saturday, October 24, 2026 · 10 AM–5 PM PT 📍AGI House · Hillsborough, CA Hosted by Google DeepMind and AGI House Apply to build → agihouse.org/events/the-goog… #Gemma #NVIDIA #Jetson #EdgeAI #PhysicalAI #AGIHouse

What if I told you there was a model that can do it all? Literally. You can use it as a VLM, vision language model, to reason over interactions in a video. You can even use it as a robot policy, as a backbone for WAM, world action models. You can use it as a physics simulator to test your policies and even generate infinite amounts of synthetic data for your robot training. But it doesn’t stop there. They just released their smallest version, the Cosmos 3 Edge, which is only 4 billion parameters, allowing you to run it on edge devices like Jetson AGX Orin or Thor. And that’s exactly what I’ll be doing today. By the end of this video, you’ll see how to run it on the Jetson AGX Orin and see how well it performs. Let’s get to it. Check out the full video here: piped.video/H_acNcH5ZZ4?is=sKaV… Try it out Cosmos 3: github.com/NVIDIA/cosmos-fra… Thank you @NVIDIARobotics for the Jetson AGX Orin!

Road to GTC Berlin: Europe’s AI Builders in the Spotlight nitter.siddz.com/i/broadcasts/1dxYlaroQ…

👀 See how startups work through real engineering challenges, live at #NVIDIAGTC Berlin. At Startup Stack-a-thon, founders and NVIDIA solutions architects will map workflows on a digital whiteboard to unpack technical bottlenecks. 🛠️ 🗓️ Wednesday, October 21 🕞 3:30 p.m. – 4:10 p.m. with prefix.dev & Haptikos Inc. 🕟 4:30 p.m. – 5:10 p.m. with Plim Robotics & Port. Join the audience and follow the thinking behind the stack. Add the sessions to your schedule: nvda.ws/4yqP2a3

We are happy to annouced @iuliaferoli just dropped a hands-on video walkthrough using the #reBotArmB601 and our free Seeed × @NVIDIARobotics DLI course. The full Physical AI lifecycle, in real code: → Chapter 1: Data collection with the reBot arm → Chapter 2: Simulation & sim-to-real in Isaac Sim → Chapter 3: VLA training & tuning (Isaac GR00T) → Chapter 4: Edge deployment on Jetson Thor → Chapter 5: Real-world picks & lessons learned This is what "Physical AI for everyone" looks like — a robotic arm, a free DLI course, and a community that's already shipping. More episodes coming on the YouTube channel, first one is live now. Watch, follow, build your own project. #SeeedStudio #reBot #NVIDIA #IsaacSim #LeRobot #AIRoboics #PhysicalAI #OpenSource #TheAIHarewarePartner

Can’t make it to #NVIDIAGTC Berlin? We’ve got you covered. 🎙 Join us for 3 days of livestreams covering the biggest moments of the week, from the keynote recap to real-world AI demos and developer stories straight from Berlin. Add the following streams to your calendar: → Mon, Oct 19 | Road to GTC: Europe Recap nvda.ws/3U61a1U → Wed, Oct 21 | GTC Berlin Day 1: Keynote Recap nvda.ws/4yOS9JN → Thu, Oct 22 | GTC Berlin Day 2: Inside the Experience nvda.ws/4zwMesL

How can robots learn dexterous manipulation from human demonstrations at scale? Excited to share CHORD: Learning Dexterous Manipulation Using Contact Wrench Guidance From Human Demonstration. CHORD learns from human demos by focusing not only on where contact happens, but how that contact moves the object through force and torque guidance. This unified contact-wrench representation carries human manipulation skills across diverse behaviors, long-horizon tasks, whole-body embodiments, and real-world hardware. We evaluated CHORD on large-scale, long-horizon, contact-rich tasks paired with human demonstrations, spanning rigid, articulated, and multi-object manipulation. At scale: * 82.12% average success across 1,831 tasks * 90.77% whole-body manipulation success * 4,739 sim-ready dexterous manipulation benchmark * Transfer to real dexterous hands Project page: nvidia-isaac.github.io/video… Tech report: nvidia-isaac.github.io/video… Code will be released soon as part of Video to Data repo github.com/nvidia-isaac/vide…, our end-to-end pipeline for converting human demonstration videos into simulation-ready assets and physics-grounded robot training data. Huge thanks to amazing contributors: @zhu_xinghao , Zixi Liu, Shalin Jain, Chenran Li, Milad Noori, Huihua Zhao, John Welsh, @michaelv03, Wei Liu, @TingwuWang , Xingye (Dennis) Da, @zhengyiluo, Vishal Kulkarni, @sNaema, @yukez, @DrJimFan, @bowenwen_me, @danfei_xu, @SohaPouya, @Dr_YanChang. #Robotics #PhysicalAI #DexterousManipulation #RobotLearning #NVIDIA

Can and should a general-purpose model serve as a robot policy? In Astranomy, we study GPT-6 Astra for general pick-and-place manipulation. Astra sets a new SOTA on RoboLab-120, but real-time robot control remains an open challenge. Full tech report: research.nvidia.com/labs/srl…

What’s New in Newton Physics for Robot Simulation | Robotics Office Hour nitter.siddz.com/i/broadcasts/1vJpPNmBm…

What’s new in Newton? See it in action. Join our Robotics Office Hours on October 14 at 11 AM PT for a live walkthrough of Newton’s latest features, including robot control and inverse kinematics. Explore how these workflows support robot learning, with highlights from the “Accelerating Robot Learning With Isaac Lab and Newton” learning path. Bring your questions for a live Q&A. 📅 Add to calendar: addevent.com/event/jhwfdnz4h…

(1/6) From @GoogleDeepmind’s AlphaGo to AlphaZero, we saw machines find moves that reached beyond what we’d imagined for them. No one taught them those strategies. They emerged from reinforcement learning: playing, failing, and trying again under a single binary task reward for success or failure. Inspired by them, we couldn’t help but wonder: what might a robot discover if we gave it that same freedom to learn? That question led us to AlphaNIST. Starting tabula rasa, with that same binary reward plus only a simple control cost, a single policy solves all 20 @NIST assembly tasks at a 98.2% success rate in simulation. More surprising still, AlphaNIST can complete 10 different tasks in a row, receiving that reward only once, after all 10 are done. Just as AlphaGo’s strategies surprised its creators, AlphaNIST discovered dexterous strategies we couldn’t have come up with ourselves. Paper: arxiv.org/abs/2610.13354 Website: alpha-nist.github.io

What will it take for AI to move beyond generating pixels and start understanding worlds? At #NVIDIAGTC Berlin, hear from @bfl_ai, @LTX, and @vesoma_ai for a conversation about the future of world models. Explore whether traditional scaling laws still apply, where foundation models end and robotics begins, and what the open ecosystem must standardize to accelerate progress. 🕒 October 22 @ 9 AM CEST 📅 Add to calendar: nvidia.com/en-eu/gtc/session…

Where physical AI meets the farm 🌱 @Aigeninc is using NVIDIA Cosmos and NVIDIA Jetson to build solar-powered robots for autonomous, chemical-free weed control, reducing reliance on herbicides and diesel-powered field passes. With NVIDIA technology, Aigen can: ✅ Train perception models with 99% synthetic data ✅ Reduce data collection and labelling to turn out effective models in hours instead of seasons Read the story: nvidia.com/en-us/case-studie…