Berkeley open-sources Specialist: 3D-printed gripper for robot data, runs on real arms
SongShuran · x · 2026-11-21
UC Berkeley open-sourced Specialist: a handheld gripper you can 3D print, plus April tags, forming a fully open-source stack to record UMI-style robot training data. Policies trained only on those handheld demonstrations run on a real robot arm.
A Rerun blog post (by Pablo Vela) highlights what makes the release special: the team used Rerun at every step of the pipeline — raw camera capture → dataset processing → end-to-end training — not just for visual inspection. This follows the approach of companies like Sunday Robotics and Generalist AI, which collect demonstrations with handheld grippers instead of teleoperated robots: faster and cheaper per demo, with complexity shifted into the offline pipeline that recovers what the hand actually did. Specialist (from Yufeng Chi, Pieter Abbeel, Koushil Sreenath, et al.), inspired by Generalist AI's grippers, is an open-source implementation of that pipeline. The post walks through how it uses Rerun and how to run it without owning any hardware.
More from Embodied
- Open-source humanoid Asimov starts wheeled variant on day 389 of development — freelerobot · 2026-11-21
- USC's Heave-Ho trains Unitree G1 humanoids to collaborate, scales 2 to 10 robots — CyberRobooo · 2026-11-21
- Wanting a full self-driving Tesla RV: futuristic Cybertruck-style camper concept — PaulYacoubian · 2026-11-21
- PhysLDM reconstructs deformable trajectories at 2.48mm precision with 78x token compression — jiqizhixin · 2026-11-21
- Open-source autonomous driving book's Chapter 4 compares Pure Pursuit, Stanley, rear-wheel feedback — 4310sy · 2026-11-21
- SAFI teaches VLA robots to stop safely mid-task, cutting takeover success to 1.3% — DM2-ND · 2026-11-21