返回全部动态

LeRobot v0.7.0:从TB级数据到提示机器人

原标题:LeRobot v0.7.0: From Terabytes to Prompting Robots

Hugging Face Blog一手来源开源质量 82

AI 摘要

Hugging Face 发布 LeRobot v0.7.0,支持直接从 Hub、Storage Bucket 或 LanceDB 流式训练近 300TB 开放机器人数据,无需先下载。新版本重建训练栈以支持 5-20B 参数的世界动作模型,并引入交互式 rollout,可在机器人运行时通过文本指令切换子任务、提问或自动导航。同时集成 NVIDIA、Black Forest Labs 等四个世界动作模型和四个新 VLA,支持远程 GPU 推理。

以上摘要由 AI 生成,可能存在误差。事实请以原文为准。

正文节选

Every robot policy starts as terabytes of demonstrations and ends as a robot following instructions. LeRobot v0.7.0 works on both ends of that road. The Hub now holds almost 300 TB of open Physical AI data in LeRobot format, with some of the largest datasets released in just the past few months. You can now train on these datasets directly from the Hub, a Storage Bucket or LanceDB, without downloading the datasets first. The training stack has been rebuilt for multi-billion-parameter world actio


发布时间:—
抓取时间:2026-11-09 20:48
来源机构:Hugging Face
阅读原文huggingface.co