The open stack for robot learning keeps maturing. This release shifts the focus from 'react' to 'predict'.
Fact
The facts: on July 7, 2026 Hugging Face released LeRobot v0.6.0. It brings policies built on world models (VLA-JEPA, FastWAM, LingBot-VA) that learn to predict future states; new VLA models; reward models to score success; six simulation benchmarks under one evaluation CLI; support for depth sensors; automatic language annotation through a VLM (a model that understands text and images); FSDP training across many GPU-equipped machines, and cloud training through Hugging Face Jobs. Primary source: Hugging Face (huggingface.co).
The core of it is the 'world model'. The robot imagines the next state and measures itself against it, instead of learning blind, by trial and error. That's a change of direction, not another changelog-style checkbox.
But the model is open - so it isn't the edge. The winner is whoever accumulates their own recordings of real movement and knows how to tie them together. The data from your robots becomes the moat. Everyone has the architecture.
Official primary sources
Hugging Face - LeRobot v0.6.0 releaseThe visual is generated code-art, no third-party images.