Hy-Embodied-0.5-VLA — LeRobot Conversion
lerobot/hy_vla_robotwin
LeRobot has released a converted version of the Hy-Embodied-0.5-VLA-RoboTwin model, achieving full forward parity with the original model. The model supports a 16D dual-arm embodiment and has been validated across multiple metrics, including task completion and action accuracy. The release highlights the model's performance on a RoboTwin quick protocol and notes limitations in expressing mobile-manipulator actions.
Full article
You are reading the complete RoboSignal summary. The publisher’s full article is available at the original source.
Read full article at sourcehuggingface.co · Opens in a new tab; source language may differ.
The release details a model conversion and validation process for a robotic policy, showing full parity between the original and converted model across various metrics. The model supports a specific 16D dual-arm embodiment and has limitations in expressing mobile-manipulator actions.
Evidence and limits
Published automatically after robotics and source-evidence checks; no manual editorial approval is recorded. Source assertions are not independently verified. Missing information remains not reported.
- Environment:
- Not reported
- Control:
- Not reported
- Data origin:
- Not reported
- dataset: Not reported
Source excerpts and review record
No manual editorial approval recorded.
Original source quotation: “Hy-Embodied-0.5-VLA — LeRobot conversion Byte-identical tensor conversion of tencent/Hy-Embodied-0.5-VLA-RoboTwin at revision bd7bba6f5934ad62293a2a34f74760c6a3ef2ff8 .”
Source E1
Original source quotation: “Model token horizon: 40 Physical action horizon: 20 Actions executed per sample: 7 Image history: 6 frames at interval 5 Action representation: relative_absolute Euler inference steps: 10”
Source E2
Original source quotation: “Validation Strict load, save_pretrained , and strict reload passed with all 1,628 tensors mapped and no missing, unexpected, duplicate, or shape-mismatched tensors.”
Source E3
Original source quotation: “On the pinned first-six-task RoboTwin quick protocol, the author wrapper scored 60/60. LeRobot scored 53/54 completed episodes; conservatively counting one native-planner hang as a failure gives 53/55 (96.36%) versus the matched author subset's 55/55, a 3.64-percentage-point delta.”
Source E4
Original source quotation: “Limitation This policy supports the released 16D dual-arm pose/gripper embodiment only. A 12D mobile-manipulator action contains base semantics that the released 20D dual-arm representation cannot express and is rejected rather than silently truncated or reinterpreted.”
Source E5
Source：LeRobot — Model repository updates · huggingface.co