Orca opens a shared stack for dexterous hand research and teleoperation
tensorqt · x · 2026-07-26
A new paper introduces orca, an open-source stack for dexterous manipulation research.
The paper argues that anthropomorphic hands are a better platform for dexterous learning than two-finger grippers, but current hand research is fragmented across ad-hoc codebases. Orca aims to unify low-level control, simulation, teleoperation, and hand retargeting behind one interface, while integrating with popular robot-learning frameworks such as lerobot.
The authors say the stack supports:
- teleoperation from consumer VR headsets
- expert demonstration collection for in-hand reorientation
- training autonomous policies with lerobot
- reproducible evaluation in a fully observable setup
They position the release as a shared, reproducible foundation for open dexterity research.
More from Embodied
- A OnePlus 12 can run Z Image Turbo and Flux.2 locally, but a 320×320 image still takes minutes — sgcego · 2026-07-26
- Jensen Huang says Optimus could help create a multi-trillion-dollar humanoid robot industry — XFreeze · 2026-07-26
- Weekly AI paper roundup covers world models, embodied AI, and self-improving agents — _akhaliq · 2026-07-26
- Alibaba and CUHK’s RynnWorld-4D predicts a robot scene’s future from one image — jiqizhixin · 2026-07-26
- Vivix A1 accepts voice, text and images mid-interaction, demo says — Scobleizer · 2026-07-26
- Vivix A1 demo shows synchronized speech, gaze, facial motion and body movement — Scobleizer · 2026-07-26