
𝗟𝗲𝗥𝗼𝗯𝗼𝘁 𝘃𝟬.𝟳.𝟬 𝗶𝘀 𝗟𝗜𝗩𝗘 🚀🤖 From terabytes of data to prompting robots! Check out all what's new in the video below and join the 50,000+ developers & 300+ organizations in the LeRobot ecosystem 🤗
✓~ Lowering the barrier to entry for robotics ~ Crafted with care by @HuggingFace 🤗 Join our discord: discord.com/invite/s3KuuzsPF…

𝗟𝗲𝗥𝗼𝗯𝗼𝘁 𝘃𝟬.𝟳.𝟬 𝗶𝘀 𝗟𝗜𝗩𝗘 🚀🤖 From terabytes of data to prompting robots! Check out all what's new in the video below and join the 50,000+ developers & 300+ organizations in the LeRobot ecosystem 🤗

Today we're open-sourcing HubEgo-100K: the most diverse egocentric dataset on 🤗 @huggingface. 100,000 hours of skilled tasks from 100,000+ different Hub contributors. Free for research and commercial use 🧵

YAM: huggingface.co/docs/lerobot/… reBot B601: huggingface.co/docs/lerobot/… Unitree G1 + SONIC: huggingface.co/docs/lerobot/…

More arms. More humanoids. 🦾 New hardware in LeRobot 0.7: - YAM from @i2rt_robotics : drive one arm or two, with gravity feed-forward computed from the arm's own model - reBot B601 from @seeedstudio: now on RobStride or Damiao motors, and a bimanual setup can mix the two - @UnitreeRobotics G1 + NVIDIA SONIC: your policy outputs a 64-D motion token, and the SONIC decoder turns it into 50 Hz targets for all 29 joints. Works with remote inference - The 23-DoF G1 joins too: same driver, one setting Yes, that's our G1 dodging a basketball 🏀 Docs in the comments.

Remote inference: huggingface.co/docs/lerobot/… Coming from async_inference? It's been replaced: pip install "lerobot[remote]" Humanoids too: with SONIC, the policy runs on the GPU and the @UnitreeRobotics G1's Jetson runs the decoder.

Run big policies on a remote GPU ⚡🤖 Your robot keeps moving while the GPU thinks. New in LeRobot 0.7: the robot computer runs cameras, motors and recording, and a policy server on a GPU machine predicts the next chunk of actions. An action buffer asks for the next chunk early, so the robot doesn't stop between chunks. Two commands: lerobot-policy-server --config_path=server.yaml lerobot-rollout --inference.type=remote ... Same machine, LAN or routed networks over Zenoh, with TLS and access control. Interactive sessions, episode recording and RTC come along, and it's tested on real robots with ACT, SmolVLA, XVLA, EVO1 and LaWAM. Docs in the comments 👇

You can benchmark an LLM from a laptop, but robot policies need real hardware / environments for testing. Yesteday, @pepijn2233 shared how @huggingface is releasing every RoboColosseum eval rollout, including failures. What a great to wrap up @corl_conf!

G0.5: huggingface.co/docs/lerobot/… LingBot-VLA 2.0: huggingface.co/docs/lerobot/… DM0.5: huggingface.co/docs/lerobot/… Wall-OSS 0.5: huggingface.co/docs/lerobot/…

Five new VLAs in LeRobot 0.7. The zoo keeps growing 🤖 - G0.5 from @GalaxeaDynamics: a 2B VLA that writes its subtask, then its actions, in one stream. Works with /subtask and /autosteer - LingBot-VLA 2.0 from @robbyant_brain: one model across embodiments (55-D action space, ~60,000 h of pretraining). Its sparse MoE runs an action chunk in 120 ms in LeRobot, down from 756 ms - DM0.5 from @Dexmal_AI: a Gemma3-4B VLM + a 680M flow-matching action expert, on standard LeRobot datasets - Wall-OSS 0.5 from @XSquareRobot: a 4B VLA that also writes text. It answers /vqa and plans its own subtasks - FineART-VLA: π0.5 with a language head, trained on FineART (more in our language post) Docs in the comments 👇

10,000 hours of open egocentric data with SOTA 3D head and hand tracking from @GroundedSI are now available 🤗 RoboCap-10k spans 21 industries in 5 continents with 619 real, paid work tasks all using 6-camera/dual IMU RoboCap hardware. Explore the dataset and research ↓


LeRobot docs: huggingface.co/docs/lerobot/… Blog: runway.com/news/company-news…

Praxis-1: an open-weight world action model for real robots 🤖 New from @runwayml and now in LeRobot. Praxis-1 learns how objects behave and how tasks unfold from large-scale video, which gives it a head start on a new robot. One checkpoint adapts with light fine-tuning or LoRA: fine-tuned on Google's RT-1 robot, which it never saw in training, it reaches 80.0% on SimplerEnv. Weights, inference and fine-tuning code are open under Apache 2.0, commercial use included, and it runs on NVIDIA from H100 to Jetson Thor. Model, blog and docs in the comments 👇

Cosmos3: huggingface.co/docs/lerobot/… FLUX 3 Action: huggingface.co/docs/lerobot/… LaWAM: huggingface.co/docs/lerobot/… Long-WAM: huggingface.co/docs/lerobot/…

Robots that imagine before they act 🔮🤖 World action models predict what happens next, then act. LeRobot 0.7 adds four of them: - Cosmos3 from @NVIDIARobotics: 16.5B Nano and 4.6B Edge, 97.9% LIBERO average. Edge runs on one GPU - FLUX 3 Action from @bfl_ai: one open 7B diffusion transformer predicts future video and actions together, with an SO-101 checkpoint - LaWAM from RLinf: imagines the future in latent space (DINOv3 features), 98.5% on LIBERO measured in LeRobot - Long-WAM from NVIDIA: looks back over 19.2 s of video history, 99.0% on LIBERO measured in LeRobot Docs in the comments 👇

Robot policies work in the lab. The real world is where they break. To close that gap, we are open-sourcing the most diverse egocentric dataset with the help of @LeRobotHF. Coming to 🤗 @huggingface on Thursday November 12th.