LeRobot v0.4.0: Super Charging OSS Robotics Learning

Why It Matters

Together these changes significantly lower the barriers to large‑scale robot learning and evaluation, speeding experiment throughput, improving dataset manageability, and broadening hardware and benchmark interoperability for researchers and builders.

Summary

Hugging Face released LeRobot v0.4.0, a major upgrade to its open‑source robotics stack that introduces LeRobotDataset v3.0 (chunked episodes and streaming to handle OXE‑scale datasets >400 GB), new VLA models (PI0.5 and GR00T N1.5), and a plugin system for easier hardware integration. The update also adds official support for LIBERO (130+ VLA tasks) and Meta‑World (50+ manipulation tasks), dataset editing tools, a modular Processor pipeline for seamless model‑to‑robot data flows, and one‑command multi‑GPU training via Accelerate. Together these changes significantly lower the barriers to large‑scale robot learning and evaluation, speeding experiment throughput, improving dataset manageability, and broadening hardware and benchmark interoperability for researchers and builders.