Pure-vision robot policy: two wrist RGBD cams + claimed "GPT6" hits near-100% success
ChongZzZhang · x · 2026-09-21
A demo (video sped 4x/12x) shows a robot manipulation policy with no base/world-frame observation and no gripper feedback — just plain visual understanding from two RGBD cameras on the same wrist. The author claims a "GPT6 as policy" directly outputs end-effector frame actions from the two cams plus instructions, with near-100% success given inference tokens and time for retries. Tasks include finding an occluded wooden box and placing it on a yellow sign, and stacking boxes. Note the "GPT6" claim is unverified.
Related event: GPT6 as Policy: Wrist-Mounted Dual Cameras Drive Zero-Shot Robot Stacking(5 posts)→
More from Embodied
- 1X targets 50,000 humanoid robots shipped by 2027, ramping to 110k/year capacity — Scobleizer · 2026-09-21
- Optimus lead says robot moves are remote-controlled; physical AI safety dwarfs LLM safety — Scobleizer · 2026-09-21
- 'Dedicated to the AI safety kids': robot demo looks 'pretty unwinnable for the human' — burny_tech · 2026-09-21
- GPT-6-controlled robot arm breaks gripper during LEGO grasping; team seeks force-aware control — paigeinsf · 2026-09-21
- Digit 4 demo raises questions: no speed specs, big backpack, possible CGI footage — MarwaEldiwiny · 2026-09-21
- Human vs Robot Fight at San Francisco's REK Event Hit 'Completely Different', Says Attendee — BLUECOW009 · 2026-09-21