Open-ENPIRE-Gripper democratizes NVIDIA's ENPIRE gripper design for diverse robot arms
chris_j_paxton · x · 2026-10-07
Developer pgeedh open-sourced Open-ENPIRE-Gripper (Apache-2.0), democratizing the compliant gripper design from NVIDIA GEAR's ENPIRE project (co-led by Jim Fan and Yuke Zhu) so it works across different robot arms.
The repo provides Universal Compliant Gripper (UCG) adapters, already mounting onto commercial Robotiq Hand-E grippers, with STL files, scripts, and docs. Rationale: in physical AI manipulation research, grasping reliability hinges on hardware contact mechanics — surface friction, passive compliance, tactile geometry — which is how ENPIRE achieved remarkable real-world grasping reliability. Endorsed by researchers like Chris Paxton.
More from Embodied
- Scale and Hugging Face built a new bimanual robot dataset for training VLA models — yacinelearning · 2026-10-07
- Hands-on with Standard Bots' BOLT bimanual arms targeting manufacturing teleop data at IROS — jonstephens85 · 2026-10-07
- Intel's Kinematic MeanFlow Cuts GR00T Action-Head Latency by up to 74% with One-Step Generation — IntelLabsChina · 2026-10-07
- ENPIRE harness automates robotics research, hitting 99% success on dexterous tasks — chris_j_paxton · 2026-10-07
- MIT leaves a coding agent alone on a digital island for 30 hours to study if machines play — phillip_isola · 2026-10-07
- Engineers 'decommission' Booster robot with a front flip, trained end-to-end in half a day — RemiCadene · 2026-10-07