Hugging Face Blog·· 2026-07-07
LeRobot v0.6.0: Imagine, Evaluate, Improve
LeRobot v0.6.0: Imagine, Evaluate, Improve
AI summary
Hugging Face released LeRobot v0.6.0, introducing world model policies VLA-JEPA, FastWAM and LingBot-VA, new VLAs GR00T N1.7, MolmoAct2, EO-1, EVO1 and Multitask DiT, and a unified reward model API with Robometer and TOPReward for the robotics learning loop.
Selection record
Threshold 60Official, first-handFirst 74Second 72
AdmittedSum of both 146 ≥ twice the threshold 120
- Source tier
- Official, first-hand; this tier's threshold is 60
- Pre-filter
- passed：LeRobot发布，含VLA、世界模型等AI技术
- Why it was chosen
- Integrates world model policies, reward model APIs and six simulation benchmarks into one toolchain, showing the current form of the open-source robotics learning loop.
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: Hugging Face Blog · huggingface.co