I wrote a fun little article about all the ways to dodge the need for real-world robot data. I think it has a cute title. sergeylevine.substack.com/p/sporks-of-agi

wait omg that was me

NVIDIA AI Releases GraspGen: A Diffusion-Based Framework for 6-DOF Grasping in Robotics NVIDIA GraspGen is an advanced diffusion-based framework designed for 6-DOF robotic grasping. By harnessing a vast simulation-generated dataset of 53 million grasps across more than 8,000

Introducing Lume, the robotic lamp. The first robot designed to fit naturally into your home and help with chores, starting with laundry folding. If you’re looking for help and want to avoid the privacy and safety concerns of humanoids in your home, pre-order now.

Modern AI is confined to the digital world. At Skild AI, we are building towards AGI for the real world, unconstrained by robot type or task — a single, omni-bodied brain. Today, we are sharing our journey, starting with early milestones, with more to come in the weeks ahead.

Got back last night from the World AI Conference in Shanghai. Megathread with photos/videos/thoughts from the conf itself + giant expo next door (ended up going back to the expo 3 times bc there were so many interesting booths) First up: robots robots robots (yes, inc Unitree)

⚠️ Due to an overwhelming number of requests, #CoRL2025 registration is temporarily closed. One presenter per accepted paper will be guaranteed registration, and registration will reopen at a later date.

𝗣𝗼𝗽𝘂𝗹𝗮𝗿 𝗼𝗽𝗶𝗻𝗶𝗼𝗻: "𝗝𝘂𝘀𝘁 𝗴𝗲𝗻𝗲𝗿𝗮𝘁𝗲 𝗺𝗼𝗿𝗲 𝘀𝗶𝗺𝘂𝗹𝗮𝘁𝗶𝗼𝗻 𝗱𝗮𝘁𝗮." After working with many 𝗿𝗼𝗯𝗼𝘁 𝗺𝗮𝗻𝗶𝗽𝘂𝗹𝗮𝘁𝗶𝗼𝗻 teams who've fallen into the simulation trap, here's what I've learned: Simulation teaches your robot to be really,

This is insanely cool, how does one get an invite The Humanoid Hub

The wait is over. 🥁 NVIDIA Isaac Sim 5.0 and Isaac Lab 2.2 are now generally available. Dive in and start to simulate, train, and validate your AI-powered robots with these new key features: ⚡ Instant GPU access via NVIDIA Brev 🤖 3D #GaussianSplatting with @NVIDIAOmniverse

Danfei Xu Bring DARPA Robotics Challenge back? Nothing forces systems research like a real robot competition

Checkout Jun Yamada's latest work on learning Value Functions for MPC! 💡Key Idea: End-to-end policies (e.g. VLA, Transformer, Diffusion Policy, etc.) do not explicitly reason about robot and world physics (dynamics, collision constraints) when generating motions. Instead, by

Looking forward to the Conference on Robot Learning CoRL2025 Diamond Sponsor talk sessions Jonghyun Choi and I will be hosting today in the main auditorium. There should be a lot of new exciting research and announcements! Session 1, Sunday Sept 28 [TODAY] 1:30 – 1:45 pm: "General purpose

Visuomotor sim2real policy training with photorealistic rendering has always been challenging. But not anymore with IsaacLab! Checkout the latest Isaac Lab whitepaper to learn how NVIDIA Isaac Lab extends GPU-native simulation into large-scale training for a whole suite of

Happy to share that I’ve been selected for MIT Technology Review Innovators Under 35 Asia Pacific 2025! MIT Technology Review 🙏 Grateful to my mentors and research collaborators at NVIDIA Robotics, CMU and beyond. This recognizes some of our work on scaling robot learning with procedural

This deserves more attention! A real-world benchmark with 30 tasks across 4 robotic platforms, designed to evaluate VLA models. Each task includes 1,000 demos for finetuning. VLAs still have a long way to go. 🤖📊

We just started hiring for 2026 summer internships! 🤖 PhD Robotics Research Internship applications at NVIDIA AI are now open: nvidia.eightfold.ai/careers?query=… We are mainly looking for candidates with prior experience and interest in robot manipulation, RL, VLAs, sim2real. Please DM

Proudest moment of my research career, Neural MP won Best Student Paper at IROS! We built a fast, reactive neural motion planner that works in the real world and ended up being used to be enable incredibly long-horizon manipulation tasks in ManipGen (mihdalal.github.io/manipgen/).

If you feel like giving up, you must read this never-before-shared story of the creator of PyTorch and ex-VP at Meta, Soumith Chintala. > from hyderabad public school, but bad at math > goes to a "tier 2" college in India, VIT in Vellore > rejected from all 12 universities for