LeRobot v0.6.0: Imagine, Evaluate, Improve
huggingface.co · 2,417 words · saved by 1 readers
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
This new release is about closing the robot learning loop: policies that imagine the future before acting, reward models that tell you when your robot succeeds, a deployment CLI that turns failures into training data, and six new simulation benchmarks to measure it all. It also brings depth sensing, VLM-powered dataset annotation, custom video encoding, cloud training on HF Jobs, and a much leaner install. TL;DR LeRobot v0.6.0 introduces world model policies (VLA-JEPA, FastWAM, LingBot-VA) that learn to imagine the future, a wave of new VLAs (GR00T N1.7, MolmoAct2, EO-1, EVO1, Multitask…
saved by
related reading
- GitHub - robotics-survey/Awesome-Robotics-Foundation-Modelsgithub.com
- State of Robot Learning, December 2025vedder.io
- How Claude Performs on Robotics Tasks \ Anthropicanthropic.com
- Explore | alphaXivalphaxiv.org
- A VLA with Open-World Generalizationpi.website
- how we accidentally solved robotics by watching 1 million hours of YouTube – atharva's blogksagar.bearblog.dev
- Causal Video Models Are Data-Efficient Robot Policy Learners | Rhoda AIrhoda.ai
- State of Vision-Language-Action (VLA) Research at ICLR 2026 – Moritz Reussmbreuss.github.io
- Pretrained to Imagine, Fine-Tuned to Act: The Rise of World-Action Models | NVIDIA Technical Blogdeveloper.nvidia.com
- Precise Manipulation with Efficient Online RLpi.website
- pistar06.pdfpi.website
- SimpleVLA-RL: Scaling VLA Training via Reinforcement Learningalphaxiv.org