Get your robot working quickly and easily with SenseCraft Robotics — sign up for 1,000 free credits to train 3-5 models! 🔗 Claim your 1,000 free credits here: #SenseCraftRobotics# #Robotics# #RoboticArm# #RobotTraining# #EmbodiedAI# #AI#

Humanoid robot company Figure is launching a gig economy platform called Index to gather robot training data from human workers wearing cameras and sensors. Check out the full story to learn more: 📸: John Koetsier

Internet video is now robot training data! 🛜 @perceptroninc, the startup from the ex-Meta team behind the Chameleon multimodal models, just released Isaac 0.5, an open-source embodied foundation model. The same weights answer questions about a video, point to objects, track them over time, report task progress, and generate the robot's actions. What it learns from perception directly shapes what it does. To make a big model run fast enough for control, they built on something called Null Experts: each token can pick how many experts it activates, so the model spends more compute on hard parts of a scene and coasts through the easy ones. → 36B-parameter sparse model, trained across 35+ robot systems and 3T multimodal tokens → 62.6 on ScreenSpot-Pro grounding vs 54.8 for the strongest Qwen3-VL run → Does it at ~8.5× lower inference cost, 26.9 TFLOP per request against 228.4 → On a chess-manipulation task, one epoch on a single episode cut action loss 10.5× Perception and action sharing one brain, with compute that flexes to the task, that's a genuinely different shape for a robot model. And it's fully open, weights, training code, inference via LeRobot. Here's the blog: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

you can collect robot training data without robots now HiFi-UMI records humans performing tasks with a handheld gripper, then uses that data to post-train policies that deploy directly on real robots Across 3 model families, it matched robot teleoperation within 3.1 percentage points Best result: 85% success on precision insertion. it's just a matter of collecting accurate human demonstrations now.

🤖 What if you could pilot a robot with a VLM that never sees a single robot training example? RoboDawn tackles exactly that question. Title: Transferring the Intelligence of VLMs to Robotic Control (RoboDawn) URL: It gives a frozen, pretrained VLM a human-intuitive interface of translation, rotation, and gripper commands, plus a handful of in-context demonstrations, and lets it directly drive a robot arm. Three things stand out. 🎮 A game-like control interface The VLM issues discrete move, rotate, and gripper commands, all defined relative to the gripper interaction point. This lets it reuse spatial manipulation knowledge it already picked up from web-scale pretraining. 📚 One demo makes a huge difference No parameter updates at all. Just a command primer plus task demonstrations as context lift success on RoboTwin 2.0 from 53.2% zero-shot to 73.6% one-shot. 🏆 It beats robot-trained policies outright With zero task-specific training, RoboDawn's zero-shot performance already surpasses policies trained on dedicated data, like π0.5 (46.0%) and LingBot-VLA (50.4%). On a real Franka robot it hits a 90% success rate. It suggests the real bottleneck may not be collecting more robot data, but designing the interface that unlocks the intelligence VLMs already have. #Robotics# #VLM#

📍World Youth Development Forum participants witness humanoid robot training in Wuhan #Robot# #Youth# #Wuhan# @AmbXieFeng @SpoxCHN_LinJian @xuejianosaka @zhang_heqing @salahzhang @consulat_de @ChenPingMFA @CGMeifangZhang

Newton 1.5 is now available, making robot training at scale faster and more reliable. This release delivers: 🔁 More parallel simulation, lower memory, selective resets 🤝 Consistent contact physics 🤖 Experimental batched GPU control 📦 Cleaner USD/MJCF imports Get started:

I just got named the 1st @StrikeRobot_ai Core Contributor. Here's what that means and why I think it matters. Strike Robot is building the SR Platform, an AI-native robotics environment. 50 contributors were picked to shape it from the ground up before public launch. Not advisors. Not investors. Builders with actual access. I got seat #1#. Here's the part most people won't see coming. As a contributor, I can build my own simulation layouts on the platform. Get familiar with the system. Create. Iterate. Then sell those layouts. The layouts get used to train robots. Think about that for a second. You're not just a user on this platform. You're producing the training data that teaches physical robots how to operate. And you get paid for it. That's the mechanic: ➜ Access SR Platform as a Core Contributor ➜ Build and test your own simulation layouts ➜ Sell your layouts to the platform ➜ Your work directly trains real robots The broader play here is a creator economy for robot training data. The people who get in now and build strong layouts are the ones who own the library when demand scales. Reps now. Revenue later.